You searched for subject:( Machine Learning).
Showing records 1 – 30 of
11287 total matches.
◁ [1] [2] [3] [4] [5] … [377] ▶
Oregon State University
1.
Vatturi, Pavan Kumar.
Rare category detection using hierarchical mean shift.
Degree: MS, Computer Science, 2009, Oregon State University
URL: http://hdl.handle.net/1957/10191 
► Many applications in surveillance, monitoring, scientific discovery, and data cleaning require the identification of anomalies. Although many methods have been developed to identify statistically significant…
(more)
▼ Many applications in surveillance, monitoring, scientific discovery, and data cleaning require the identification of anomalies. Although many methods have been developed to identify statistically significant anomalies, a more difficult task is to identify anomalies that are both interesting and statistically significant. Category detection is an emerging area of
machine learning that can address this issue using a "human-in-the-loop" approach. In this interactive setting, the algorithm asks the user to label a query data point under an existing category or declare the query data point to belong to a previously undiscovered category. The goal of category detection is to discover all the categories in the data in as few queries as possible. In a data set with imbalanced categories, the main challenge is in identifying the rare categories or anomalies; hence, the task is often referred to as rare category detection.
We present a new approach to rare category detection using a hierarchical mean shift procedure. In our approach, a hierarchy is created by repeatedly applying mean shift with increasing bandwidth on the entire data set. This hierarchy allows us to identify anomalies in the data set at different scales, which are then posed as queries to the user. The main advantage of this methodology over existing approaches is that it does not require any knowledge of the data set properties such as the total number of classes or the prior probabilities of the classes. Results on real-world data sets show that our hierarchical mean shift approach performs consistently better than previous techniques.
Advisors/Committee Members: Wong, Weng-Keen (advisor), Fern, Alan (committee member).
Subjects/Keywords: machine learning; Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Vatturi, P. K. (2009). Rare category detection using hierarchical mean shift. (Masters Thesis). Oregon State University. Retrieved from http://hdl.handle.net/1957/10191
Chicago Manual of Style (16th Edition):
Vatturi, Pavan Kumar. “Rare category detection using hierarchical mean shift.” 2009. Masters Thesis, Oregon State University. Accessed March 04, 2021.
http://hdl.handle.net/1957/10191.
MLA Handbook (7th Edition):
Vatturi, Pavan Kumar. “Rare category detection using hierarchical mean shift.” 2009. Web. 04 Mar 2021.
Vancouver:
Vatturi PK. Rare category detection using hierarchical mean shift. [Internet] [Masters thesis]. Oregon State University; 2009. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/1957/10191.
Council of Science Editors:
Vatturi PK. Rare category detection using hierarchical mean shift. [Masters Thesis]. Oregon State University; 2009. Available from: http://hdl.handle.net/1957/10191
Oregon State University
2.
Bao, Xinlong.
Applying machine learning for prediction, recommendation, and integration.
Degree: PhD, Computer Science, 2009, Oregon State University
URL: http://hdl.handle.net/1957/12549
► This dissertation explores the idea of applying machine learning technologies to help computer users find information and better organize electronic resources, by presenting the research…
(more)
▼ This dissertation explores the idea of applying
machine learning technologies to help computer users find information and better organize electronic resources, by presenting the research work conducted in the following three applications: FolderPredictor, Stacking Recommendation Engines, and Integrating
Learning and Reasoning.
FolderPredictor is an intelligent desktop software tool that helps the user quickly locate files on the computer. It predicts the file folder that the user will access next by applying
machine learning algorithms to the user's file access history. The predicted folders are presented in existing Windows GUIs, so that the user's cost for
learning new interactions is minimized. Multiple prediction algorithms are introduced and their performance is examined in two user studies.
Recommender systems are one of the most popular means of assisting internet users in finding useful online information. The second part of this dissertation presents a novel way of building hybrid recommender systems by applying the idea of Stacking from ensemble
learning. Properties of the input users/items, called runtime metrics, are employed as additional meta features to improve performance. The resulting system, called STREAM, outperforms each component engine and a static linear hybrid system in a movie recommendation problem.
Many desktop assistant systems help users better organize their electronic resources by incorporating
machine learning components (e.g., classifiers) to make intelligent predictions. The last part of this dissertation addresses the problem of how to improve the performance of these
learning components, by integrating
learning and reasoning through Markov logic. Through an inference engine called the PCE, multiple classifiers are integrated via a process called relational co-training that improves the performance of each classifier based on information propagated from other classifiers.
Advisors/Committee Members: Dietterich, Thomas G. (advisor), Bergman, Lawrence (committee member).
Subjects/Keywords: machine learning; Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Bao, X. (2009). Applying machine learning for prediction, recommendation, and integration. (Doctoral Dissertation). Oregon State University. Retrieved from http://hdl.handle.net/1957/12549
Chicago Manual of Style (16th Edition):
Bao, Xinlong. “Applying machine learning for prediction, recommendation, and integration.” 2009. Doctoral Dissertation, Oregon State University. Accessed March 04, 2021.
http://hdl.handle.net/1957/12549.
MLA Handbook (7th Edition):
Bao, Xinlong. “Applying machine learning for prediction, recommendation, and integration.” 2009. Web. 04 Mar 2021.
Vancouver:
Bao X. Applying machine learning for prediction, recommendation, and integration. [Internet] [Doctoral dissertation]. Oregon State University; 2009. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/1957/12549.
Council of Science Editors:
Bao X. Applying machine learning for prediction, recommendation, and integration. [Doctoral Dissertation]. Oregon State University; 2009. Available from: http://hdl.handle.net/1957/12549
Oregon State University
3.
Liu, Liping.
Machine Learning Methods for Computational Sustainability.
Degree: PhD, Computer Science, 2016, Oregon State University
URL: http://hdl.handle.net/1957/59159
► Maintaining the sustainability of the earth’s ecosystems has attracted much attention as these ecosystems are facing more and more pressure from human activities. Machine learning…
(more)
▼ Maintaining the sustainability of the earth’s ecosystems has attracted much attention as these
ecosystems are facing more and more pressure from human activities.
Machine learning can
play an important role in promoting sustainability as a large amount of data is being collected
from ecosystems. There are at least three important and representative issues in the study of
sustainability: detecting the presence of species, modeling the distribution of species, and protecting
endangered species. For these three issues, this thesis selects three typical problems as
the main focus and studies these problems with different
machine learning techniques. Specifically,
this thesis investigates the problem of detecting bird species from bird song recordings,
the problem of modeling migrating birds at the population level, and the problem of designing a
conservation area for an endangered species. First, this thesis models the problem of bird song
classification as a weakly-supervised
learning problem and develops a probabilistic classification
model for the
learning problem. The thesis also analyzes the learnability of the superset label
learning problem to determine conditions under which one can learn a good classifier from the
training data. Second, the thesis models bird migration with a probabilistic graphical model at
the population level using a Collective Graphical Model (CGM). The thesis proposes a Gaussian
approximation to significantly improve the inference efficiency of the model. Theoretical results
show that the proposed Gaussian approximation is correct and can be calculated efficiently.
Third, the thesis studies a typical reserve design problem with a novel formulation of transductive
classification. Then the thesis solves the formulation with two optimization algorithms. The
learning techniques in this thesis are general and can also be applied to many other
machine
learning problems.
Advisors/Committee Members: Dietterich, Thomas (advisor), Fern, Xiaoli Z. (committee member).
Subjects/Keywords: machine learning; Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Liu, L. (2016). Machine Learning Methods for Computational Sustainability. (Doctoral Dissertation). Oregon State University. Retrieved from http://hdl.handle.net/1957/59159
Chicago Manual of Style (16th Edition):
Liu, Liping. “Machine Learning Methods for Computational Sustainability.” 2016. Doctoral Dissertation, Oregon State University. Accessed March 04, 2021.
http://hdl.handle.net/1957/59159.
MLA Handbook (7th Edition):
Liu, Liping. “Machine Learning Methods for Computational Sustainability.” 2016. Web. 04 Mar 2021.
Vancouver:
Liu L. Machine Learning Methods for Computational Sustainability. [Internet] [Doctoral dissertation]. Oregon State University; 2016. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/1957/59159.
Council of Science Editors:
Liu L. Machine Learning Methods for Computational Sustainability. [Doctoral Dissertation]. Oregon State University; 2016. Available from: http://hdl.handle.net/1957/59159
Oregon State University
4.
Hooper, Samuel.
Spatial and Temporal Dynamics of Broad-scale Predictive Models : Influences of Scale.
Degree: MS, Geography, 2017, Oregon State University
URL: http://hdl.handle.net/1957/60148
► Developing accurate predictive distribution models requires adequately representing relevant spatial and temporal scales, as these scales are ultimately reflective of the relationships between distributions and…
(more)
▼ Developing accurate predictive distribution models requires adequately representing relevant spatial and temporal scales, as these scales are ultimately reflective of the relationships between distributions and influential environmental conditions. In this research, we considered both spatial and temporal scale and the influence each has on predicting broad-scale distributions of two disparate but related phenomena: land cover and bird distributions. Employing
machine-
learning algorithms, we first developed land cover time series datasets covering all of California, Oregon, and Washington with a model that simultaneously reflects local-scale heterogeneity and broad-scale homogeneity. We then used these and other land cover time series datasets to investigate the effects of temporal resolution on species distribution models.
In the second chapter, we focused on the importance of accurately representing the spatial scale of relationships between predictors and a response variable for broad-scale predictive models. Using both a novel
machine-
learning algorithm and a novel predictor dataset, we developed dense time series forest canopy cover (FCC) and impervious surface cover (ISC) datasets at a 30-meter spatial resolution for all of California, Oregon, and Washington. To develop both datasets, we employed a spatial ensemble modeling method using a population of locally defined and spatially overlapping decision trees, making it both appropriate at continental-scales and sensitive to local variation in predictor-response relationships. Our predictor variables were products of LandTrendr, a tool for developing time series images and derivatives from the Landsat archive. To develop the most accurate time series of FCC and ISC, we first tested two model parameters, sample size and estimator size. Using the best-performing configuration of each, we then compared our models with locally defined estimators to bagged decision trees, the most comparable model with globally defined estimators. Using the best-performing models and LandTrendr imagery, we developed yearly FCC and ISC maps, spanning 1990-2012. To test the temporal extensibility of our models, we compared our predicted 2011 maps to 2011 maps from the National Land Cover Database. We found that model performance for both FCC and ISC decreased with increasing estimator size and that models with locally defined estimators outperformed bagged decision trees. We also found that our models performed well when extending learned predictor-response relationships to predict 2011 FCC and ISC distributions. These results, in concert with several novel byproducts of the models that we developed, demonstrate that representing local-scale spatial relationships is critical to producing accurate broad-scale distribution models.
In the third chapter, we investigated the influence of temporal scale on an avian species distribution model (SDM) by comparing models developed with different temporal resolutions of land cover predictor data. We expressed temporal resolution as the time…
Advisors/Committee Members: Kennedy, Robert E. (advisor), Robinson, Douglas (committee member).
Subjects/Keywords: machine learning; Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Hooper, S. (2017). Spatial and Temporal Dynamics of Broad-scale Predictive Models : Influences of Scale. (Masters Thesis). Oregon State University. Retrieved from http://hdl.handle.net/1957/60148
Chicago Manual of Style (16th Edition):
Hooper, Samuel. “Spatial and Temporal Dynamics of Broad-scale Predictive Models : Influences of Scale.” 2017. Masters Thesis, Oregon State University. Accessed March 04, 2021.
http://hdl.handle.net/1957/60148.
MLA Handbook (7th Edition):
Hooper, Samuel. “Spatial and Temporal Dynamics of Broad-scale Predictive Models : Influences of Scale.” 2017. Web. 04 Mar 2021.
Vancouver:
Hooper S. Spatial and Temporal Dynamics of Broad-scale Predictive Models : Influences of Scale. [Internet] [Masters thesis]. Oregon State University; 2017. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/1957/60148.
Council of Science Editors:
Hooper S. Spatial and Temporal Dynamics of Broad-scale Predictive Models : Influences of Scale. [Masters Thesis]. Oregon State University; 2017. Available from: http://hdl.handle.net/1957/60148
Texas A&M University
5.
Kapale, Anurag.
An Automated Framework To Generate End-to-End Machine Learning Pipelines.
Degree: MS, Computer Science, 2019, Texas A&M University
URL: http://hdl.handle.net/1969.1/188756
► The recent developments in machine learning have shown its applicability in numerous real-world applications. However, building an optimal machine learning pipeline requires considerable knowledge and…
(more)
▼ The recent developments in
machine learning have shown its applicability in numerous real-world applications. However, building an optimal
machine learning pipeline requires considerable knowledge and experience in data science. To address this problem, many automated
machine learning (AutoML) frameworks have been proposed. However, most of the existing AutoML frameworks treat the pipeline generation as a black-box optimization problem. Thus, failing to incorporate basic heuristics and human intuition. Furthermore, most of these frameworks provide very basic or no feature engineering abilities. To tackle these challenges, in this thesis, we propose an automated framework to generate end-to-end
machine learning pipelines. By survey of 100s of Kaggle kernels and extensive experimentation, we finalized a set of heuristics which enhances the pipeline optimization problem. We also implemented a system to automate feature engineering, which could generate 100s of features to produce better representation of the data.
Additionally, the framework provides interpretations about why certain models and features were selected by the system. This would help the users to further improve the pipeline. Finally, our experimentation shows consistent performance across various datasets.
Advisors/Committee Members: Hu, Xia (advisor), Mortazavi, Bobak (committee member), Shen, Yang (committee member).
Subjects/Keywords: Machine Learning; Automated Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Kapale, A. (2019). An Automated Framework To Generate End-to-End Machine Learning Pipelines. (Masters Thesis). Texas A&M University. Retrieved from http://hdl.handle.net/1969.1/188756
Chicago Manual of Style (16th Edition):
Kapale, Anurag. “An Automated Framework To Generate End-to-End Machine Learning Pipelines.” 2019. Masters Thesis, Texas A&M University. Accessed March 04, 2021.
http://hdl.handle.net/1969.1/188756.
MLA Handbook (7th Edition):
Kapale, Anurag. “An Automated Framework To Generate End-to-End Machine Learning Pipelines.” 2019. Web. 04 Mar 2021.
Vancouver:
Kapale A. An Automated Framework To Generate End-to-End Machine Learning Pipelines. [Internet] [Masters thesis]. Texas A&M University; 2019. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/1969.1/188756.
Council of Science Editors:
Kapale A. An Automated Framework To Generate End-to-End Machine Learning Pipelines. [Masters Thesis]. Texas A&M University; 2019. Available from: http://hdl.handle.net/1969.1/188756
6.
Ramezanpanah, Zahra.
Bi-lateral interaction between a humanoid robot and a human in mixed reality : Interaction bi latérale entre un robot humanoïde et un humain assistée par la réalité mixte.
Degree: Docteur es, Robotique, 2020, université Paris-Saclay
URL: http://www.theses.fr/2020UPASG039
► Cette thèse peut être divisée en deux parties: la reconnaissance des actions et la reconnaissance des émotions. Chaque partie se fait selon deux méthodes, la…
(more)
▼ Cette thèse peut être divisée en deux parties: la reconnaissance des actions et la reconnaissance des émotions. Chaque partie se fait selon deux méthodes, la méthode classique de Machine Learning et le réseau profond. Dans la section Reconnaissance des actions, nous avons d'abord défini un descripteur local basé sur la LMA, pour décrire les mouvements. LMA est un algorithme pour décrire un mouvement en utilisant ses quatre composants: le corps, l'espace, la forme et l'effort. Le seul objectif de cette partie étant la reconnaissance des gestes, seuls les trois premiers facteurs ont été utilisés. L'algorithme DTW, est implémenté pour trouver les similitudes des courbes obtenues à partir des vecteurs descripteurs obtenus par la méthode LMA. Enfin SVM, l'algorithme est utilisé pour former et classer les données. Dans la deuxième partie de cette section, nous avons construit un nouveau descripteur basé sur les coordonnées géométriques de différentes parties du corps pour présenter un mouvement. Pour ce faire, en plus des distances entre le centre de la hanche et les autres articulations du corps et les changements des angles de quaternion dans le temps, nous définissons les triangles formés par les différentes parties du corps et calculons leur surface. Nous calculons également l'aire de la seule frontière 3D conforme autour de toutes les articulations du corps. À la fin, nous ajoutons la vitesse de l'articulation différente dans le descripteur proposé. Nous avons utilisé LSTM pour évaluer ce descripteur. Dans la deuxième partie de cette thèse, nous avons d'abord présenté un module de niveau supérieur pour identifier les sentiments intérieurs des êtres humains en observant leurs mouvements corporels. Afin de définir un descripteur robuste, deux méthodes sont mises en œuvre: La première méthode est la LMA, qui en ajoutant le facteur «Effort» est devenue un descripteur robuste, qui décrit un mouvement et l'état dans lequel il a été effectué. De plus, le second sur est basé sur un ensemble de caractéristiques spatio-temporelles. Dans la suite de cette section, un pipeline de reconnaissance des mouvements expressifs est proposé afin de reconnaître les émotions des personnes à travers leurs gestes par l'utilisation de méthodes d'apprentissage automatique. Une étude comparative est faite entre ces 2 méthodes afin de choisir la meilleure. La deuxième partie de cette partie consiste en une étude statistique basée sur la perception humaine afin d'évaluer le système de reconnaissance ainsi que le descripteur de mouvement proposé.
This thesis can be divided into two parts: action recognition and emotion recognition. Each part is done in two method, classic method of Machine Learning and deep network. In the Action Recognition section, we first defined a local descriptor based on the LMA, to describe the movements. LMA is an algorithm to describe a motion by using its four components: Body, Space, Shape and Effort. Since the only goal in this part is gesture recognition, only the first three factors have been used. The DTW,…
Advisors/Committee Members: Mallem, Malik (thesis director).
Subjects/Keywords: Machine Learning; Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Ramezanpanah, Z. (2020). Bi-lateral interaction between a humanoid robot and a human in mixed reality : Interaction bi latérale entre un robot humanoïde et un humain assistée par la réalité mixte. (Doctoral Dissertation). université Paris-Saclay. Retrieved from http://www.theses.fr/2020UPASG039
Chicago Manual of Style (16th Edition):
Ramezanpanah, Zahra. “Bi-lateral interaction between a humanoid robot and a human in mixed reality : Interaction bi latérale entre un robot humanoïde et un humain assistée par la réalité mixte.” 2020. Doctoral Dissertation, université Paris-Saclay. Accessed March 04, 2021.
http://www.theses.fr/2020UPASG039.
MLA Handbook (7th Edition):
Ramezanpanah, Zahra. “Bi-lateral interaction between a humanoid robot and a human in mixed reality : Interaction bi latérale entre un robot humanoïde et un humain assistée par la réalité mixte.” 2020. Web. 04 Mar 2021.
Vancouver:
Ramezanpanah Z. Bi-lateral interaction between a humanoid robot and a human in mixed reality : Interaction bi latérale entre un robot humanoïde et un humain assistée par la réalité mixte. [Internet] [Doctoral dissertation]. université Paris-Saclay; 2020. [cited 2021 Mar 04].
Available from: http://www.theses.fr/2020UPASG039.
Council of Science Editors:
Ramezanpanah Z. Bi-lateral interaction between a humanoid robot and a human in mixed reality : Interaction bi latérale entre un robot humanoïde et un humain assistée par la réalité mixte. [Doctoral Dissertation]. université Paris-Saclay; 2020. Available from: http://www.theses.fr/2020UPASG039
Rutgers University
7.
Imtiaz, Hafiz, 1986-.
Decentralized differentially private algorithms for matrix and tensor factorization.
Degree: PhD, Electrical and Computer Engineering, 2020, Rutgers University
URL: https://rucore.libraries.rutgers.edu/rutgers-lib/62938/
► Many applications of machine learning, such as human health research, involve processing private or sensitive information. Privacy concerns may impose significant hurdles to collaboration in…
(more)
▼ Many applications of machine learning, such as human health research, involve processing private or sensitive information. Privacy concerns may impose significant hurdles to collaboration in scenarios where there are multiple sites holding data and the goal is to estimate properties jointly across all datasets. Conventional differentially private decentralized algorithms can provide strong privacy guarantees. However, the utility/accuracy of the joint estimates may be poor when the datasets at each site are small. In this work, we propose a new framework, Correlation Assisted Private Estimation (CAPE), for designing privacy-preserving decentralized algorithms with much better accuracy guarantees in an honest-but-curious model. We show that CAPE can be employed in a range of decentralized computations common in machine learning problems.
We note that matrix and tensor factorizations are key components of many decentralized processing pipelines that involve joint subspace learning. In this work, we focus on principal component analysis, independent component analysis, canonical correlation analysis and orthogonal tensor decomposition. Conventional decentralized differentially private factorization schemes suffer from excessive noise, which leads to sub-optimal subspace/feature learning. We demonstrate that the CAPE framework fits perfectly in these problems and can be employed to remedy the excessive noise issue. More specifically, we develop decentralized algorithms for these matrix and tensor factorization problems and show that, under certain conditions, these algorithms can achieve the same utility as a centralized algorithm using all datasets across sites.
Finally, we employ our CAPE framework to propose an algorithm for solving generalized optimization problems in decentralized settings. We provide a tighter characterization of the functional mechanism and propose modifications such that it can be incorporated in the CAPE framework. Our proposed decentralized functional mechanism is specifically suited for privacy-preserving computation of virtually any differentialble and continuous cost function in the decentralized setting.
For all of our proposed algorithms, we present empirical results to demonstrate that our proposed algorithms outperform existing state-of-the-art algorithms and can be competitive with non-private algorithms in many scenarios of interest. Our results indicate that meaningful privacy can be attained without losing much performance by the virtue of better algorithm design.
Advisors/Committee Members: Sarwate, Anand (chair), Dana, Kristin (internal member), Soljanin, Emina (internal member), Yates, Roy (internal member), Calhoun, Vince (outside member), School of Graduate Studies.
Subjects/Keywords: Decentralized machine learning; Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Imtiaz, Hafiz, 1. (2020). Decentralized differentially private algorithms for matrix and tensor factorization. (Doctoral Dissertation). Rutgers University. Retrieved from https://rucore.libraries.rutgers.edu/rutgers-lib/62938/
Chicago Manual of Style (16th Edition):
Imtiaz, Hafiz, 1986-. “Decentralized differentially private algorithms for matrix and tensor factorization.” 2020. Doctoral Dissertation, Rutgers University. Accessed March 04, 2021.
https://rucore.libraries.rutgers.edu/rutgers-lib/62938/.
MLA Handbook (7th Edition):
Imtiaz, Hafiz, 1986-. “Decentralized differentially private algorithms for matrix and tensor factorization.” 2020. Web. 04 Mar 2021.
Vancouver:
Imtiaz, Hafiz 1. Decentralized differentially private algorithms for matrix and tensor factorization. [Internet] [Doctoral dissertation]. Rutgers University; 2020. [cited 2021 Mar 04].
Available from: https://rucore.libraries.rutgers.edu/rutgers-lib/62938/.
Council of Science Editors:
Imtiaz, Hafiz 1. Decentralized differentially private algorithms for matrix and tensor factorization. [Doctoral Dissertation]. Rutgers University; 2020. Available from: https://rucore.libraries.rutgers.edu/rutgers-lib/62938/
University of California – San Diego
8.
Gallagher, Patrick W.
Operator Theory for Analysis of Convex Optimization Methods in Machine Learning.
Degree: Cognitive Science, 2014, University of California – San Diego
URL: http://www.escholarship.org/uc/item/153375qt
► As machine learning has more closely interacted with optimization, the concept of convexity has loomed large. Two properties beyond simple convexity have received particularly close…
(more)
▼ As machine learning has more closely interacted with optimization, the concept of convexity has loomed large. Two properties beyond simple convexity have received particularly close attention: strong smoothness and strong convexity. These properties (and their relatives) underlie machine learning analyses from convergence rates to generalization bounds – they are central and fundamental. This thesis takes as its focus properties from operator theory that, in specific instances, relate to broadened conceptions of convexity, strong smoothness, and strong convexity. Some of the properties we consider coincide with strong smoothness and strong convexity in some settings, but represent broadenings of these concepts in other situations of interest. Our intention throughout is to take an approach that balances theoretical generality with ease of use and subsequent extension. Through this approach we establish a framework, novel in its scope of application, in which a single analysis serves to recover standard convergence rates (typically established via a variety of separate arguments) for convex optimization methods prominent in machine learning. The framework is based on a perspective in which the iterative update for each convex optimization method is regarded as the application of some operator. We establish a collection of correspondences, novel in its comprehensiveness, that exist between "contractivity- type'' properties of the iterative update operator and "monotonicity-type'' properties of the associated displacement operator. We call particular attention to the comparison between the broader range of properties that we discuss and the more restricted range considered in the contemporary literature, demonstrating as well the relationship between the broader and narrower range. In support of our discussion of these property correspondences and the optimization method analyses based on them, we relate operator theory concepts that may be unfamiliar to a machine learning audience to more familiar concepts from convex analysis. In addition to grounding our discussion of operator theory, this turns out to provide a fresh perspective on many touchstone concepts from convex analysis
Subjects/Keywords: Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Gallagher, P. W. (2014). Operator Theory for Analysis of Convex Optimization Methods in Machine Learning. (Thesis). University of California – San Diego. Retrieved from http://www.escholarship.org/uc/item/153375qt
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Gallagher, Patrick W. “Operator Theory for Analysis of Convex Optimization Methods in Machine Learning.” 2014. Thesis, University of California – San Diego. Accessed March 04, 2021.
http://www.escholarship.org/uc/item/153375qt.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Gallagher, Patrick W. “Operator Theory for Analysis of Convex Optimization Methods in Machine Learning.” 2014. Web. 04 Mar 2021.
Vancouver:
Gallagher PW. Operator Theory for Analysis of Convex Optimization Methods in Machine Learning. [Internet] [Thesis]. University of California – San Diego; 2014. [cited 2021 Mar 04].
Available from: http://www.escholarship.org/uc/item/153375qt.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Gallagher PW. Operator Theory for Analysis of Convex Optimization Methods in Machine Learning. [Thesis]. University of California – San Diego; 2014. Available from: http://www.escholarship.org/uc/item/153375qt
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
University of Oxford
9.
Cobb, Adam Derek.
The practicalities of scaling Bayesian neural networks to real-world applications.
Degree: PhD, 2020, University of Oxford
URL: http://ora.ox.ac.uk/objects/uuid:4b738b70-28bc-4545-86a6-6078861e7d13
;
https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.800130
► In this work, I will focus on ways in which we can build machine learning models that appropriately account for uncertainty, whether with computationally cheap…
(more)
▼ In this work, I will focus on ways in which we can build machine learning models that appropriately account for uncertainty, whether with computationally cheap estimates or with more expensive and reliable ones. In particular, I will explore how we can model distributions with Bayesian neural networks and how we can manipulate them depending on the task. The two main techniques for performing inference in Bayesian neural networks are variational inference and Markov chain Monte Carlo. I will look into the advantages and disadvantages of both methods and apply them to real-world problems. The emphasis is on how to achieve calibrated uncertainty estimates without compromising scalability. One contribution of this work is to offer a new method for implementing Bayesian neural networks within the framework of Bayesian decision theory, where Bayesian decision theory is important in all decision-making applications. A further contribution is developing sampling techniques that provide more reliable uncertainties, especially over data that lie outside the training distribution. Finally I also introduce a method for using Bayesian neural networks in an astrophysical application where it is vital that uncertainties are calibrated appropriately for the task.
Subjects/Keywords: Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Cobb, A. D. (2020). The practicalities of scaling Bayesian neural networks to real-world applications. (Doctoral Dissertation). University of Oxford. Retrieved from http://ora.ox.ac.uk/objects/uuid:4b738b70-28bc-4545-86a6-6078861e7d13 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.800130
Chicago Manual of Style (16th Edition):
Cobb, Adam Derek. “The practicalities of scaling Bayesian neural networks to real-world applications.” 2020. Doctoral Dissertation, University of Oxford. Accessed March 04, 2021.
http://ora.ox.ac.uk/objects/uuid:4b738b70-28bc-4545-86a6-6078861e7d13 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.800130.
MLA Handbook (7th Edition):
Cobb, Adam Derek. “The practicalities of scaling Bayesian neural networks to real-world applications.” 2020. Web. 04 Mar 2021.
Vancouver:
Cobb AD. The practicalities of scaling Bayesian neural networks to real-world applications. [Internet] [Doctoral dissertation]. University of Oxford; 2020. [cited 2021 Mar 04].
Available from: http://ora.ox.ac.uk/objects/uuid:4b738b70-28bc-4545-86a6-6078861e7d13 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.800130.
Council of Science Editors:
Cobb AD. The practicalities of scaling Bayesian neural networks to real-world applications. [Doctoral Dissertation]. University of Oxford; 2020. Available from: http://ora.ox.ac.uk/objects/uuid:4b738b70-28bc-4545-86a6-6078861e7d13 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.800130
University of Oxford
10.
Assael, Ioannis Alexandros.
Deep learning for communication : emergence, recognition and synthesis.
Degree: PhD, 2019, University of Oxford
URL: http://ora.ox.ac.uk/objects/uuid:794592eb-d957-49c4-8d46-c4eb19bd0125
;
https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.799986
► Human intelligence is a social phenomenon tightly coupled to the act and process of communication. Ever since the early prehistoric period, humans have been able…
(more)
▼ Human intelligence is a social phenomenon tightly coupled to the act and process of communication. Ever since the early prehistoric period, humans have been able to communicate amongst themselves at an unprecedented and unparalleled level compared to all other living species. Communication led humans to develop media such as the spoken and written word to effectively convey the meanings of concrete and abstract concepts, and still today a substantial part of human life is spent communicating and sharing information. The scientific study of communication began in Classical Greece with the work of Aristotle and was to evolve through time into the work on information theory by Claude E. Shannon. This work proposes three novel methods for studying the processes of emergence, recognition, synthesis and enhancement of communication, using recent advances in deep learning. The first method investigates the emergence of communication among agents, and introduces a differentiable way of learning communication protocols. The second studies speech recognition in visual verbal communication, and for the first time solves sentence-level lipreading with deep neural networks trained end-to-end. The third and final method proposes a meta-learning approach for sample efficient verbal communication via text-to-speech synthesis. This thesis advances deep learning in these areas, and defines the premises for the creation of novel technologies for the greater good of society.
Subjects/Keywords: Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Assael, I. A. (2019). Deep learning for communication : emergence, recognition and synthesis. (Doctoral Dissertation). University of Oxford. Retrieved from http://ora.ox.ac.uk/objects/uuid:794592eb-d957-49c4-8d46-c4eb19bd0125 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.799986
Chicago Manual of Style (16th Edition):
Assael, Ioannis Alexandros. “Deep learning for communication : emergence, recognition and synthesis.” 2019. Doctoral Dissertation, University of Oxford. Accessed March 04, 2021.
http://ora.ox.ac.uk/objects/uuid:794592eb-d957-49c4-8d46-c4eb19bd0125 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.799986.
MLA Handbook (7th Edition):
Assael, Ioannis Alexandros. “Deep learning for communication : emergence, recognition and synthesis.” 2019. Web. 04 Mar 2021.
Vancouver:
Assael IA. Deep learning for communication : emergence, recognition and synthesis. [Internet] [Doctoral dissertation]. University of Oxford; 2019. [cited 2021 Mar 04].
Available from: http://ora.ox.ac.uk/objects/uuid:794592eb-d957-49c4-8d46-c4eb19bd0125 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.799986.
Council of Science Editors:
Assael IA. Deep learning for communication : emergence, recognition and synthesis. [Doctoral Dissertation]. University of Oxford; 2019. Available from: http://ora.ox.ac.uk/objects/uuid:794592eb-d957-49c4-8d46-c4eb19bd0125 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.799986
11.
Khorshidi, Alireza.
Methods and Theories in Atomistic Reaction
Engineering.
Degree: School of Engineering, 2018, Brown University
URL: https://repository.library.brown.edu/studio/item/bdr:792658/
► Electronic-structure calculations have provided us with in-depth understanding of chemical and physical phenomena in atomic and molecular scales. In particular, density functional theory has been…
(more)
▼ Electronic-structure calculations have provided us
with in-depth understanding of chemical and physical phenomena in
atomic and molecular scales. In particular, density functional
theory has been extensively used to complement experimental
observations, also to provide insights and mechanisms not
accessible by experiment in a wide range of scientific research
areas. There is however still a significant gap between the scale
of realistic experiments and the scale affordable by
quantum-mechanics approaches. This dissertation is devoted to two
possible approaches in order to narrow this gap. The first approach
is based on
machine learning, where quantum-mechanics potential
energy surface is learned by agnostic function approximators. The
function approximator can then be used for calculation of systems
having larger size in longer time-scale simulations, with typically
orders-of-magnitude faster speed as compared to quantum-mechanics
methods. In this dissertation we address several possible
frameworks for developing
machine-
learning potentials and force
fields. In particular, we discuss our Atomistic
Machine-
learning
Package (Amp) that we developed during the course of this thesis
preparation. Amp is an open-source package in python with fortran
counterparts to accelerate computationally-intensive calculations.
For the purpose of illustration, we then show how
machine-
learning
calculations inside Amp can accelerate stable cluster searches,
transition state searches, or can be coupled with quantum mechanics
in quantum-mechanics/
machine-
learning type of hybridization
schemes. As a second approach to save the number of required
quantum-mechanical calculations, we discuss possible theories
inspired from continuum mechanics that can qualitatively predict
the results of electronic-structure calculations, without the need
to do extra quantum mechanics. We illustrate such
continuum-mechanics approaches within the context of the effect of
mechanical strain on the energy of adsorption of gas-phase chemical
species on catalysts' surfaces. The methods and theories
discussed in this thesis exemplify possible approaches to narrow
the gap between quantum-mechanics scale and real experimental
scale, which seems to be of urgent need to be addressed for the
future progress in the science of physical chemistry.
Advisors/Committee Members: Peterson, Andrew A. (Advisor), Guduru, Pradeep R. (Reader), Rubenstein, Brenda M. (Reader).
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Khorshidi, A. (2018). Methods and Theories in Atomistic Reaction
Engineering. (Thesis). Brown University. Retrieved from https://repository.library.brown.edu/studio/item/bdr:792658/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Khorshidi, Alireza. “Methods and Theories in Atomistic Reaction
Engineering.” 2018. Thesis, Brown University. Accessed March 04, 2021.
https://repository.library.brown.edu/studio/item/bdr:792658/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Khorshidi, Alireza. “Methods and Theories in Atomistic Reaction
Engineering.” 2018. Web. 04 Mar 2021.
Vancouver:
Khorshidi A. Methods and Theories in Atomistic Reaction
Engineering. [Internet] [Thesis]. Brown University; 2018. [cited 2021 Mar 04].
Available from: https://repository.library.brown.edu/studio/item/bdr:792658/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Khorshidi A. Methods and Theories in Atomistic Reaction
Engineering. [Thesis]. Brown University; 2018. Available from: https://repository.library.brown.edu/studio/item/bdr:792658/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
12.
Lee, Seungjoon.
Statistical Learning Tools for Information Fusion in
Computational Fluid Dynamics.
Degree: Department of Applied Mathematics, 2017, Brown University
URL: https://repository.library.brown.edu/studio/item/bdr:733405/
► For more than a decade, remarkable scientific progress in computational fluid dynamics (CFD) has been achieved via powerful collection of tools for exascale simulations, data-driven…
(more)
▼ For more than a decade, remarkable scientific progress
in computational fluid dynamics (CFD) has been achieved via
powerful collection of tools for exascale simulations, data-driven
approaches, and
machine-
learning (or statistical
learning)
techniques that enable efficient simulations of multiscale and
multiphysics problems. With the prospect of exascale simulations in
the next decade, it is clear that new flexible tools and
specialized algorithms are required to take advantage of such
unprecedented computing environment. For example, in the DOE ASCAC
subcommittee report, it is stated ``new algorithms will need to be
designed to optimize not only for floating-point performance and
accuracy, but also to minimize associated data movement, power, and
energy cost". Furthermore, robust and fault-resilient algorithms
are increasingly attracting attention in exascale simulations
against expected and repeated software or hardware error. Hence,
new algorithms for exascale simulations inherently require the
seamless integration of robustness, resilience, correctness, and
efficiency. Computational fluid dynamics has been developed
following different pathways to address for complex flow physics
problems. Many researches have focused on building up their own
model to describe and address multi-resolution and multi-physics
applications. Based on the recent introduction of data-driven
algorithms via
machine-
learning techniques in computer simulations,
computational fluid dynamics must rely also on data-driven
approaches as much as on the anticipated improvements in computer
hardware. In data science, data from various heterogeneous sources
can be used effectively to accelerate simulations via multiple
fidelity information fusion without additional complex models,
equations, and extra state variables. This could result in
establishing a new paradigm of multifidelity simulations in
computational fluid dynamics. In this thesis, we address these new
requirements for new algorithms in exascale simulations and
introduce a novel framework including fault-resilient, robust, and
efficient algorithms via multiple fidelity information fusion
realized via statistical
learning tools. This achievement addresses
the new capability that statistical
learning techniques can bring
to traditional scientific computing algorithms. This thesis
proposes two possible directions of a next generation of
computational frameworks for exascale simulations. The first
direction addresses resilience via information fusion with
auxiliary data. In exascale simulations, if each processor can
share some global information about the simulation from a coarse,
limited accuracy but relatively costless auxiliary simulator we can
effectively fill-in the missing spatial data at the required times
by a statistical
learning technique based on multi-level Gaussian
process regression, on-the-fly. The second direction addresses
efficiency via adaptive projective time integration. The
aforementioned auxiliary data provide additional information about
dynamics-informed…
Advisors/Committee Members: Karniadakis, George (Advisor), Kevrekidis, Ioannis (Advisor), Maxey, Martin (Reader), Perdikaris, Paris (Reader).
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Lee, S. (2017). Statistical Learning Tools for Information Fusion in
Computational Fluid Dynamics. (Thesis). Brown University. Retrieved from https://repository.library.brown.edu/studio/item/bdr:733405/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Lee, Seungjoon. “Statistical Learning Tools for Information Fusion in
Computational Fluid Dynamics.” 2017. Thesis, Brown University. Accessed March 04, 2021.
https://repository.library.brown.edu/studio/item/bdr:733405/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Lee, Seungjoon. “Statistical Learning Tools for Information Fusion in
Computational Fluid Dynamics.” 2017. Web. 04 Mar 2021.
Vancouver:
Lee S. Statistical Learning Tools for Information Fusion in
Computational Fluid Dynamics. [Internet] [Thesis]. Brown University; 2017. [cited 2021 Mar 04].
Available from: https://repository.library.brown.edu/studio/item/bdr:733405/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Lee S. Statistical Learning Tools for Information Fusion in
Computational Fluid Dynamics. [Thesis]. Brown University; 2017. Available from: https://repository.library.brown.edu/studio/item/bdr:733405/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
13.
Brawner, Stephen Andrew.
Algorithms for the Personalization of AI for Robots and the
Smart Home.
Degree: Department of Computer Science, 2018, Brown University
URL: https://repository.library.brown.edu/studio/item/bdr:792905/
► Just as an interconnected-computerized world has produced large amounts of data resulting in exciting challenges for machine learning, connected households with robots and smart devices…
(more)
▼ Just as an interconnected-computerized world has
produced large amounts of data resulting in exciting challenges for
machine learning, connected households with robots and smart
devices will provide developers with an opportunity to build
technologies that learn from personalized household data. However,
there exists a dilemma. When limited data is available for a user,
for example when they initially procure a new smart device or
robot, there will be a substantial burden placed on that user to
personalize it to their household by the learner. At the outset,
applying predictions learned from a general population to a user
will provide better predictive success. But as the amount of data
provided by the user increases, intelligent methods should choose
predictions more heavily weighted by the individuals examples. This
work investigated three problems to find algorithms that learn from
both the general population and specialize to the human individual.
We developed a solution to reduce the interactive burden when
telling a robot how to organize a kitchen by applying a
context-aware recommender system. Also, using the paradigm of
trigger-action programming made popular by IFTTT, we sought to
improve the programming experience by
learning to predict the
creation of programs from the user's history. Finally we developed
several methods to personalize grounding natural language to these
trigger-action programs. In a smart home where a user can describe
to an intelligent home automated system rules or programs they
desire to be created, their utterances are highly context
dependent. Multiple users may use similar utterances to mean
different things. We present several methods that personalize the
machine translation of these utterances to smart home programs.
This work presents several problems that show that
learning
algorithms that learn from both a general population and from
personalized interactions will perform better than either
learning
approach alone.
Advisors/Committee Members: Charniak, Eugene (Reader), Littman, Michael L. (Advisor), Cakmak, Maya (Reader).
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Brawner, S. A. (2018). Algorithms for the Personalization of AI for Robots and the
Smart Home. (Thesis). Brown University. Retrieved from https://repository.library.brown.edu/studio/item/bdr:792905/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Brawner, Stephen Andrew. “Algorithms for the Personalization of AI for Robots and the
Smart Home.” 2018. Thesis, Brown University. Accessed March 04, 2021.
https://repository.library.brown.edu/studio/item/bdr:792905/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Brawner, Stephen Andrew. “Algorithms for the Personalization of AI for Robots and the
Smart Home.” 2018. Web. 04 Mar 2021.
Vancouver:
Brawner SA. Algorithms for the Personalization of AI for Robots and the
Smart Home. [Internet] [Thesis]. Brown University; 2018. [cited 2021 Mar 04].
Available from: https://repository.library.brown.edu/studio/item/bdr:792905/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Brawner SA. Algorithms for the Personalization of AI for Robots and the
Smart Home. [Thesis]. Brown University; 2018. Available from: https://repository.library.brown.edu/studio/item/bdr:792905/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
14.
Wang, Yinong.
Decoding hindlimb kinematics from primate motor cortex using
long short-term memory recurrent neural networks.
Degree: Biomedical Engineering, 2018, Brown University
URL: https://repository.library.brown.edu/studio/item/bdr:792720/
► Recent machine learning techniques have become a powerful tool in a variety of tasks, including neural decoding. Artificial neural network models, particularly recurrent models, can…
(more)
▼ Recent
machine learning techniques have become a
powerful tool in a variety of tasks, including neural decoding.
Artificial neural network models, particularly recurrent models,
can leverage the temporal evolution of neural ensemble activity to
decode complex motor and sensory signals. Using single-unit
recordings from microelectrode arrays implanted in the leg area of
primary motor cortex in non-human primates, we decoded the
positions and angles of hindlimb joints during a locomotion task
using a long short-term memory (LSTM) network. The LSTM decoder
significantly improved decoding over traditional filtering methods,
such as linear filtering techniques. However, marginal improvements
over other
machine learning and latent state-space methods were
observed.
Advisors/Committee Members: Borton, David (Advisor), Truccolo, Wilson (Advisor), Silverman, Harvey (Reader).
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Wang, Y. (2018). Decoding hindlimb kinematics from primate motor cortex using
long short-term memory recurrent neural networks. (Thesis). Brown University. Retrieved from https://repository.library.brown.edu/studio/item/bdr:792720/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Wang, Yinong. “Decoding hindlimb kinematics from primate motor cortex using
long short-term memory recurrent neural networks.” 2018. Thesis, Brown University. Accessed March 04, 2021.
https://repository.library.brown.edu/studio/item/bdr:792720/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Wang, Yinong. “Decoding hindlimb kinematics from primate motor cortex using
long short-term memory recurrent neural networks.” 2018. Web. 04 Mar 2021.
Vancouver:
Wang Y. Decoding hindlimb kinematics from primate motor cortex using
long short-term memory recurrent neural networks. [Internet] [Thesis]. Brown University; 2018. [cited 2021 Mar 04].
Available from: https://repository.library.brown.edu/studio/item/bdr:792720/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Wang Y. Decoding hindlimb kinematics from primate motor cortex using
long short-term memory recurrent neural networks. [Thesis]. Brown University; 2018. Available from: https://repository.library.brown.edu/studio/item/bdr:792720/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
15.
Patil, Prerna.
Multi-Fidelity Simulation Algorithm and its Application to
Flow over a Cylinder.
Degree: School of Engineering, 2017, Brown University
URL: https://repository.library.brown.edu/studio/item/bdr:733471/
► We tackle the classical problem of predicting the relation between of C_L , C_D and C_P vs Reynolds number for flow over cylinder using the…
(more)
▼ We tackle the classical problem of predicting the
relation between of C_L , C_D and C_P vs Reynolds number for flow
over cylinder using the multi-fidelity framework. The stochastic
response surface is obtained by implementing the auto-regressive
stochastic modeling (Kennedy and O’Hagan, 2000) and Gaussian
process regression to combine data from variable levels of
fidelity. In particular, we predict the lift, drag and pressure
coefficients where codes with multiple levels of fidelity are
available. We correlate data at each of these levels and build the
surrogate model using co-kriging technique. The deficient physics
of the low-fidelity model is explored by examining the
cross-correlation between the low-fidelity and high-fidelity
models. The proposed framework ultimately intends to meld
computational accuracy of the expensive high fidelity with the
computational cost of the inexpensive low-fidelity.
Advisors/Committee Members: Karniadakis, George (Advisor).
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Patil, P. (2017). Multi-Fidelity Simulation Algorithm and its Application to
Flow over a Cylinder. (Thesis). Brown University. Retrieved from https://repository.library.brown.edu/studio/item/bdr:733471/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Patil, Prerna. “Multi-Fidelity Simulation Algorithm and its Application to
Flow over a Cylinder.” 2017. Thesis, Brown University. Accessed March 04, 2021.
https://repository.library.brown.edu/studio/item/bdr:733471/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Patil, Prerna. “Multi-Fidelity Simulation Algorithm and its Application to
Flow over a Cylinder.” 2017. Web. 04 Mar 2021.
Vancouver:
Patil P. Multi-Fidelity Simulation Algorithm and its Application to
Flow over a Cylinder. [Internet] [Thesis]. Brown University; 2017. [cited 2021 Mar 04].
Available from: https://repository.library.brown.edu/studio/item/bdr:733471/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Patil P. Multi-Fidelity Simulation Algorithm and its Application to
Flow over a Cylinder. [Thesis]. Brown University; 2017. Available from: https://repository.library.brown.edu/studio/item/bdr:733471/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
16.
Loper, Jackson Hoy.
Theory and Computation for Modern Probabilistic
Models.
Degree: Department of Applied Mathematics, 2017, Brown University
URL: https://repository.library.brown.edu/studio/item/bdr:733424/
► Modern probabilistic models involve computation and analysis in very high-dimensional spaces. Here we explore several of ways in which analysis of problems high dimensional spaces…
(more)
▼ Modern probabilistic models involve computation and
analysis in very high-dimensional spaces. Here we explore several
of ways in which analysis of problems high dimensional spaces can
be made more tractable by various reductions. In particular, we
focus on finite approximations for sampling point processes,
particle methods for sampling high-dimensional distributions,
situations in which hitting times for brownian motion in high
dimensional space take on particularly simple forms, and certain
maximal characteristics on the infinite-dimensional space of
couplings of two random variables with fixed marginal
distributions.
Advisors/Committee Members: Geman, Stuart (Advisor), Ramanan, Kavita (Reader), Harrison, Matthew (Reader).
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Loper, J. H. (2017). Theory and Computation for Modern Probabilistic
Models. (Thesis). Brown University. Retrieved from https://repository.library.brown.edu/studio/item/bdr:733424/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Loper, Jackson Hoy. “Theory and Computation for Modern Probabilistic
Models.” 2017. Thesis, Brown University. Accessed March 04, 2021.
https://repository.library.brown.edu/studio/item/bdr:733424/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Loper, Jackson Hoy. “Theory and Computation for Modern Probabilistic
Models.” 2017. Web. 04 Mar 2021.
Vancouver:
Loper JH. Theory and Computation for Modern Probabilistic
Models. [Internet] [Thesis]. Brown University; 2017. [cited 2021 Mar 04].
Available from: https://repository.library.brown.edu/studio/item/bdr:733424/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Loper JH. Theory and Computation for Modern Probabilistic
Models. [Thesis]. Brown University; 2017. Available from: https://repository.library.brown.edu/studio/item/bdr:733424/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
17.
Chua, Jeroen.
Probabilistic Scene Grammars: A General-Purpose Framework
For Scene Understanding.
Degree: Department of Computer Science, 2017, Brown University
URL: https://repository.library.brown.edu/studio/item/bdr:792615/
► We propose a general-purpose probabilistic framework for scene understanding tasks. We show that several classical scene understanding tasks can be modeled and addressed under a…
(more)
▼ We propose a general-purpose probabilistic framework
for scene understanding tasks. We show that several classical scene
understanding tasks can be modeled and addressed under a common
representation, approximate inference scheme, and
learning
algorithm. We refer to this approach as the Probabilistic Scene
Grammar (PSG) framework. The PSG framework models scenes using
probabilistic grammars which capture relationships between objects
in terms of compositional rules that provide important contextual
cues for inference with ambiguous data. We show how to represent
the distribution defined by a probabilistic grammar using a factor
graph. We also show how to estimate the parameters of a grammar
using an approximate version of Expectation-Maximization, and
describe an approximate inference scheme using Loopy Belief
Propagation with an efficient message-passing scheme. Inference
with Loopy Belief Propagation naturally combines bottom-up and
top-down contextual information and leads to a robust algorithm for
aggregating evidence. To demonstrate the generality of the
approach, we evaluate the PSG framework on the scene understanding
tasks of contour detection, face localization, and binary image
segmentation. The results of the PSG framework are competitive with
algorithms specialized for these scene understanding
tasks.
Advisors/Committee Members: Felzenszwalb, Pedro (Advisor), Geman, Stuart (Reader), Sudderth, Erik (Reader).
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Chua, J. (2017). Probabilistic Scene Grammars: A General-Purpose Framework
For Scene Understanding. (Thesis). Brown University. Retrieved from https://repository.library.brown.edu/studio/item/bdr:792615/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Chua, Jeroen. “Probabilistic Scene Grammars: A General-Purpose Framework
For Scene Understanding.” 2017. Thesis, Brown University. Accessed March 04, 2021.
https://repository.library.brown.edu/studio/item/bdr:792615/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Chua, Jeroen. “Probabilistic Scene Grammars: A General-Purpose Framework
For Scene Understanding.” 2017. Web. 04 Mar 2021.
Vancouver:
Chua J. Probabilistic Scene Grammars: A General-Purpose Framework
For Scene Understanding. [Internet] [Thesis]. Brown University; 2017. [cited 2021 Mar 04].
Available from: https://repository.library.brown.edu/studio/item/bdr:792615/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Chua J. Probabilistic Scene Grammars: A General-Purpose Framework
For Scene Understanding. [Thesis]. Brown University; 2017. Available from: https://repository.library.brown.edu/studio/item/bdr:792615/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
18.
Ren, Zhile.
Semantic Three-Dimensional Understanding of Dynamic
Scenes.
Degree: Department of Computer Science, 2018, Brown University
URL: https://repository.library.brown.edu/studio/item/bdr:792891/
► We develop new representations and algorithms for three-dimensional (3D) scene understanding from images and videos. To model cluttered indoor scenes, we introduce object descriptors that…
(more)
▼ We develop new representations and algorithms for
three-dimensional (3D) scene understanding from images and videos.
To model cluttered indoor scenes, we introduce object descriptors
that account for camera viewpoint, and use structured
learning to
perform 3D object detection and room layout prediction. We further
boost accuracy by using latent support surfaces to capture style
variations of objects and help detect small objects. In outdoor
environments, we incorporate semantic segmentation in a cascaded
prediction framework to more accurately model the 3D scene flow of
moving objects. We first propose a cloud of oriented gradient (COG)
descriptor that links the 2D appearance and 3D pose of object
categories, and thus accurately models how perspective projection
affects perceived image boundaries. We also propose a Manhattan
voxel representation which better captures room layout geometry.
Effective classification rules are learned via a structured
prediction framework. Contextual relationships among categories and
layout are captured via a cascade of classifiers. Furthermore, we
design algorithms that use latent support surfaces to better
represent the 3D appearance of large objects, and provide
contextual cues to improve the detection of small objects. Our
model is learned solely from annotated RGB-D images, but
nevertheless its performance substantially exceeds the
state-of-the-art on the SUN RGB-D database. We then focus on
outdoor scene flow prediction. Many existing approaches use
superpixels for regularization. We instead assume that scenes
consist of foreground objects rigidly moving in front of a static
background, and use semantic cues to produce pixel-accurate scene
flow estimates. Our cascaded classification framework accurately
models scenes by iteratively refining semantic segmentation masks,
stereo correspondences, 3D rigid motion estimates, and optical flow
fields. Our method has state-of-the-art performance on the
challenging KITTI autonomous driving benchmark.
Advisors/Committee Members: Sudderth, Erik (Advisor), Felzenszwalb, Pedro (Reader), Tompkin, James (Reader).
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Ren, Z. (2018). Semantic Three-Dimensional Understanding of Dynamic
Scenes. (Thesis). Brown University. Retrieved from https://repository.library.brown.edu/studio/item/bdr:792891/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Ren, Zhile. “Semantic Three-Dimensional Understanding of Dynamic
Scenes.” 2018. Thesis, Brown University. Accessed March 04, 2021.
https://repository.library.brown.edu/studio/item/bdr:792891/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Ren, Zhile. “Semantic Three-Dimensional Understanding of Dynamic
Scenes.” 2018. Web. 04 Mar 2021.
Vancouver:
Ren Z. Semantic Three-Dimensional Understanding of Dynamic
Scenes. [Internet] [Thesis]. Brown University; 2018. [cited 2021 Mar 04].
Available from: https://repository.library.brown.edu/studio/item/bdr:792891/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Ren Z. Semantic Three-Dimensional Understanding of Dynamic
Scenes. [Thesis]. Brown University; 2018. Available from: https://repository.library.brown.edu/studio/item/bdr:792891/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
19.
Tang, Yu-Hang.
Multiscale and Mesoscopic Modeling of Soft Matter and
Biophysical Systems Using High Performance Computing and Machine
Learning.
Degree: Department of Applied Mathematics, 2017, Brown University
URL: https://repository.library.brown.edu/studio/item/bdr:792611/
► This dissertation is composed around the subject of multiscale modeling of soft matter and biophysical systems with applications using large-scale computations. Specifically, it is expanded…
(more)
▼ This dissertation is composed around the
subject of
multiscale modeling of soft matter and biophysical systems with
applications using large-scale computations. Specifically, it is
expanded on three fronts: 1) Development of high-performance
simulators and computational frameworks. On this front, I will
discuss the design of three sets of software. The first one is an
accelerated parallel particle simulator, which features many
algorithmic innovations for harnessing the massively parallel
threading architecture of general purpose graphics processing
units. The second one is an ultrafast coarse-grained molecular
dynamics simulator, which enables the simulation of an entire human
red blood cell at protein resolution using a single computer
workstation. This is realized by a novel algorithm that allows
neighbor search in a sparse 3D space in linear time.The third one
is a generic framework that utilizes the concept of meshless
interpolation to faciliate the implementation of parallel
concurrently coupled multiscale simulations. 2) Construction of
mesoscopic models for amphiphilic and thermo-responsive polymers
and their applications to large-scale mesoscopic simulations. This
is manifested in a detailed study of the non-equilibrium dynamics
of thermo-responsive polymers. One of the most interesting findings
is that a thermo-responsive polymer membrane may invert its layered
structure without actually rotating any of its composing molecules.
3) Data-driven algorithms for
learning complex interatomic force
fields. Here I have focus on a specific aspect of this field, i.e.
the design of feature vectors that can efficiently and accurately
quantify the similarity between atomistic configurations. To
achieve this, a kernel minisum approach is proposed as a robust and
efficient replacement of the principal component analysis
algorithm. A set of quadrature rules and parameters are also
proposed for constructing a smoothed density field that allows
either inner product- or norm-based comparison of structural
similarity.
Advisors/Committee Members: Karniadakis, George (Advisor), Maxey, Martin (Reader), Baker, Nathan (Reader).
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Tang, Y. (2017). Multiscale and Mesoscopic Modeling of Soft Matter and
Biophysical Systems Using High Performance Computing and Machine
Learning. (Thesis). Brown University. Retrieved from https://repository.library.brown.edu/studio/item/bdr:792611/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Tang, Yu-Hang. “Multiscale and Mesoscopic Modeling of Soft Matter and
Biophysical Systems Using High Performance Computing and Machine
Learning.” 2017. Thesis, Brown University. Accessed March 04, 2021.
https://repository.library.brown.edu/studio/item/bdr:792611/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Tang, Yu-Hang. “Multiscale and Mesoscopic Modeling of Soft Matter and
Biophysical Systems Using High Performance Computing and Machine
Learning.” 2017. Web. 04 Mar 2021.
Vancouver:
Tang Y. Multiscale and Mesoscopic Modeling of Soft Matter and
Biophysical Systems Using High Performance Computing and Machine
Learning. [Internet] [Thesis]. Brown University; 2017. [cited 2021 Mar 04].
Available from: https://repository.library.brown.edu/studio/item/bdr:792611/.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Tang Y. Multiscale and Mesoscopic Modeling of Soft Matter and
Biophysical Systems Using High Performance Computing and Machine
Learning. [Thesis]. Brown University; 2017. Available from: https://repository.library.brown.edu/studio/item/bdr:792611/
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
University of Waikato
20.
Han, Zhimeng.
Smoothing in Probability Estimation Trees
.
Degree: 2011, University of Waikato
URL: http://hdl.handle.net/10289/5701
► Classification learning is a type of supervised machine learning technique that uses a classification model (e.g. decision tree) to predict unknown class labels for previously…
(more)
▼ Classification
learning is a type of supervised
machine learning technique that uses a classification model (e.g. decision tree) to predict unknown class labels for previously unseen instances. In many applications it can be very useful to additionally obtain class probabilities for the different class labels. Decision trees that yield these probabilities are also called probability estimation trees (PETs). Smoothing is a technique used to improve the probability estimates. There are several existing smoothing methods, such as the Laplace correction, M-Estimate smoothing and M-Branch smoothing. Smoothing does not just apply to PETs. In the field of text compression, PPM in particular, smoothing methods play a important role. This thesis migrates smoothing methods from text compression to PETs. The newly migrated methods in PETs are compared with the best of the existing smoothing methods considered in this thesis under different experiment setups. Unpruned, pruned and bagged trees are considered in the experiments. The main finding is that the PPM-based methods yield the best probability estimate when used with bagged trees, but not when used with individual (pruned or unpruned) trees.
Advisors/Committee Members: Frank, Eibe (advisor).
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Han, Z. (2011). Smoothing in Probability Estimation Trees
. (Masters Thesis). University of Waikato. Retrieved from http://hdl.handle.net/10289/5701
Chicago Manual of Style (16th Edition):
Han, Zhimeng. “Smoothing in Probability Estimation Trees
.” 2011. Masters Thesis, University of Waikato. Accessed March 04, 2021.
http://hdl.handle.net/10289/5701.
MLA Handbook (7th Edition):
Han, Zhimeng. “Smoothing in Probability Estimation Trees
.” 2011. Web. 04 Mar 2021.
Vancouver:
Han Z. Smoothing in Probability Estimation Trees
. [Internet] [Masters thesis]. University of Waikato; 2011. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/10289/5701.
Council of Science Editors:
Han Z. Smoothing in Probability Estimation Trees
. [Masters Thesis]. University of Waikato; 2011. Available from: http://hdl.handle.net/10289/5701
Wake Forest University
21.
Patel, Udita.
Performance Analysis of Parallel Support Vector Machines on a MapReduce Architecture.
Degree: 2016, Wake Forest University
URL: http://hdl.handle.net/10339/59315
► The quantity of electronic data available for analysis has grown exponentially with the rapid development of the World Wide Web, the Internet of Things, and…
(more)
▼ The quantity of electronic data available for analysis has grown exponentially with the rapid development of the World Wide Web, the Internet of Things, and other digital technologies. As a result, data mining and machine learning algorithms face computational complexity issues when applied to real world datasets. Support Vector Machines (SVM) are powerful classification and regression tools but their computational requirements increase rapidly as the number of training examples increases. To address this problem, several parallel MapReduce based implementations of SVMs have been proposed. These implementation have in common that they decompose a large-scale multi-class problem to a number of relatively smaller subproblems by dividing the data into multiple partitions which can be processed in parallel; however, these approaches use different aggregation and combination strategies to form the final model.
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Patel, U. (2016). Performance Analysis of Parallel Support Vector Machines on a MapReduce Architecture. (Thesis). Wake Forest University. Retrieved from http://hdl.handle.net/10339/59315
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Patel, Udita. “Performance Analysis of Parallel Support Vector Machines on a MapReduce Architecture.” 2016. Thesis, Wake Forest University. Accessed March 04, 2021.
http://hdl.handle.net/10339/59315.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Patel, Udita. “Performance Analysis of Parallel Support Vector Machines on a MapReduce Architecture.” 2016. Web. 04 Mar 2021.
Vancouver:
Patel U. Performance Analysis of Parallel Support Vector Machines on a MapReduce Architecture. [Internet] [Thesis]. Wake Forest University; 2016. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/10339/59315.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Patel U. Performance Analysis of Parallel Support Vector Machines on a MapReduce Architecture. [Thesis]. Wake Forest University; 2016. Available from: http://hdl.handle.net/10339/59315
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Oregon State University
22.
Das, Shubhomoy.
Incorporating User Feedback into Machine Learning Systems.
Degree: PhD, 2017, Oregon State University
URL: http://hdl.handle.net/1957/61580
► Although machine learning systems are often effective in real-world applications, there are situations in which they can be even better when provided with some degree…
(more)
▼ Although
machine learning systems are often effective in real-world applications, there are situations in which they can be even better when provided with some degree of end user feedback. This is especially true when the
machine learning system needs to customize itself to the end user's preferences, such as in a recommender system, an email classifier or an anomaly detector.
This thesis explores two directions in incorporating end user feedback to
machine learning systems. First, I introduce an algorithm that incorporates feature feedback in a semi-supervised text classification setting. Feature feedback goes beyond instance-label feedback by allowing end users to indicate which feature-value combinations are predictive of the class label. In order to incorporate feature feedback in a semi-supervised setting, I develop a Locally Weighted Logistic Regression algorithm that uses a similarity metric combining information from the user's feature feedback and information based on label diffusion on the unlabeled data.
Second, I explore the use of instance-level feedback to anomaly detection algorithms. Anomaly detectors commonly return a list of the top outliers in the data. Although these outliers are statistically unusual, some are uninteresting to a user as the internal statistical model may not necessarily be aligned with the user's semantic notion of an anomaly. I present an algorithm that can increase the number of true anomalies presented to the user if a limited amount of instances are labeled as anomalous or nominal.
Advisors/Committee Members: Wong, Weng-Keen (advisor), Dietterich, Thomas (committee member).
Subjects/Keywords: Machine Learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Das, S. (2017). Incorporating User Feedback into Machine Learning Systems. (Doctoral Dissertation). Oregon State University. Retrieved from http://hdl.handle.net/1957/61580
Chicago Manual of Style (16th Edition):
Das, Shubhomoy. “Incorporating User Feedback into Machine Learning Systems.” 2017. Doctoral Dissertation, Oregon State University. Accessed March 04, 2021.
http://hdl.handle.net/1957/61580.
MLA Handbook (7th Edition):
Das, Shubhomoy. “Incorporating User Feedback into Machine Learning Systems.” 2017. Web. 04 Mar 2021.
Vancouver:
Das S. Incorporating User Feedback into Machine Learning Systems. [Internet] [Doctoral dissertation]. Oregon State University; 2017. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/1957/61580.
Council of Science Editors:
Das S. Incorporating User Feedback into Machine Learning Systems. [Doctoral Dissertation]. Oregon State University; 2017. Available from: http://hdl.handle.net/1957/61580
23.
Masci, Jonathan.
Advances in deep learning for vision, with applications to
industrial inspection: classification, segmentation and
morphological extensions.
Degree: 2014, Università della Svizzera italiana
URL: http://doc.rero.ch/record/210177
► Learning features for object detection and recognition with deep learning has received increasing attention in the past several years and recently attained widespread popularity. In…
(more)
▼ Learning features for object detection and recognition
with deep
learning has received increasing attention in the past
several years and recently attained widespread popularity. In this
PhD thesis we investigate its applications to the automatic surface
inspection system of our industrial partner ArcelorMittal, for
classification and segmentation problems. Currently employed
algorithms, in fact, use fixed feature extractors which are hard to
tune and require extensive prior-knowledge. Our work, instead,
focuses on learnable systems that can be used to improve
recognition and detection without requiring hard to obtain
task-specific domain knowledge. For image classification we propose
extensions to max-pooling convolutional networks, so that they can
be applied to solve the general defect classification problem via a
new pooling and feature encoding schemes. State-of-the-art deep
learning algorithms for object detection/segmentation have reached
outstanding performance given high-quality annotated data.
Unfortunately, they do not meet the required processing speeds of
steel industry. We propose an architecture that does not suffer the
same computational bottleneck (1500-fold speed-up) while retaining
equal performance. To further advance the field we study the
learning of morphological operators, largely used in industry. Only
few attempts have been proposed in the literature, but no approach
has ever considered the problem in its generality because of its
hard formulation. We tackle it from a different perspective and
introduce a learnable framework which seamlessly integrates
morphological operators; hence bringing these powerful tools to
deep
learning for the first time. Re-engineering an industrial
system requires time. In order to deliver an immediate return we
investigate metric
learning problems to boost performance of
currently used features. Our multimodal similarity sensitive
hashing model scales well to web-scale datasets and, thanks to the
binary representation, requires little storage and involves a cheap
distance computation. It outperforms previous state-of-the-art
approaches without requiring additional resources.
Advisors/Committee Members: Schmidhuber, Jürgen (Dir.).
Subjects/Keywords: Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Masci, J. (2014). Advances in deep learning for vision, with applications to
industrial inspection: classification, segmentation and
morphological extensions. (Thesis). Università della Svizzera italiana. Retrieved from http://doc.rero.ch/record/210177
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Masci, Jonathan. “Advances in deep learning for vision, with applications to
industrial inspection: classification, segmentation and
morphological extensions.” 2014. Thesis, Università della Svizzera italiana. Accessed March 04, 2021.
http://doc.rero.ch/record/210177.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Masci, Jonathan. “Advances in deep learning for vision, with applications to
industrial inspection: classification, segmentation and
morphological extensions.” 2014. Web. 04 Mar 2021.
Vancouver:
Masci J. Advances in deep learning for vision, with applications to
industrial inspection: classification, segmentation and
morphological extensions. [Internet] [Thesis]. Università della Svizzera italiana; 2014. [cited 2021 Mar 04].
Available from: http://doc.rero.ch/record/210177.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Masci J. Advances in deep learning for vision, with applications to
industrial inspection: classification, segmentation and
morphological extensions. [Thesis]. Università della Svizzera italiana; 2014. Available from: http://doc.rero.ch/record/210177
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Università della Svizzera italiana
24.
Bacchelli, Alberto.
Mining unstructured software data.
Degree: 2013, Università della Svizzera italiana
URL: http://doc.rero.ch/record/203066
► Our thesis is that the analysis of unstructured data supports software understanding and evolution analysis, and complements the data mined from structured sources. To this…
(more)
▼ Our thesis is that the analysis of unstructured data
supports software understanding and evolution analysis, and
complements the data mined from structured sources. To this aim, we
implemented the necessary toolset and investigated methods for
exploring, exposing, and exploiting unstructured data.To validate
our thesis, we focused on development email data. We found two main
challenges in using it to support program comprehension and
software development: The disconnection between emails and code
artifacts and the noisy and mixed-language nature of email content.
We tackle these challenges proposing novel approaches. First, we
devise lightweight techniques for linking email data to code
artifacts. We use these techniques for creating a tool to support
program comprehension with email data, and to create a new set of
email based metrics to improve existing defect prediction
approaches. Subsequently, we devise techniques for giving a
structure to the content of email and we use this structure to
conduct novel software analyses to support program comprehension.
In this dissertation we show that unstructured data, in the form of
development emails, is a valuable addition to structured data and,
if correctly mined, can be used successfully to support software
engineering activities.
Advisors/Committee Members: Michele (Dir.).
Subjects/Keywords: Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Bacchelli, A. (2013). Mining unstructured software data. (Thesis). Università della Svizzera italiana. Retrieved from http://doc.rero.ch/record/203066
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Bacchelli, Alberto. “Mining unstructured software data.” 2013. Thesis, Università della Svizzera italiana. Accessed March 04, 2021.
http://doc.rero.ch/record/203066.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Bacchelli, Alberto. “Mining unstructured software data.” 2013. Web. 04 Mar 2021.
Vancouver:
Bacchelli A. Mining unstructured software data. [Internet] [Thesis]. Università della Svizzera italiana; 2013. [cited 2021 Mar 04].
Available from: http://doc.rero.ch/record/203066.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Bacchelli A. Mining unstructured software data. [Thesis]. Università della Svizzera italiana; 2013. Available from: http://doc.rero.ch/record/203066
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
University of Oxford
25.
McLeod, Mark.
Optimizing Bayesian optimization.
Degree: PhD, 2018, University of Oxford
URL: http://ora.ox.ac.uk/objects/uuid:c35f26ba-07ec-4830-ac37-39b37d36a8b3
;
https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.786122
► We are concerned primarily with improving the practical applicability of Bayesian optimization. We make contributions in three key areas. We develop an intuitive online stopping…
(more)
▼ We are concerned primarily with improving the practical applicability of Bayesian optimization. We make contributions in three key areas. We develop an intuitive online stopping criterion, allowing only as many steps as necessary to achieve the desired target to be taken. By combining this with intelligent online switching between acquisition functions and pure local optimization we are also able to substantially improve convergence to the local minimum associated with our final solution. In cases where a continuum of reduced cost, but also reduced accuracy, evaluations are available we develop a Bayesian Optimization acquisition function to select both the location and cost of each evaluation. We achieve this with lower overheads than previous methods, translating to a real increase in performance. Part of this improvement is achieved by way of a new, more efficient, method for generating support points to sample the minimum of a Gaussian process. Further, in the case that the reduced cost estimates are unbiased we show that a practical solution cannot exist in most cases without also taking into consideration both computational overheads and a restriction on available resources. Given this knowledge we then develop a method which provides a viable solution in this setting. Finally, we outline practical implementation details for Bayesian optimization which allow substantial reductions in the overhead costs without changing the theoretical properties of optimization. This is primarily achieved by use of adaptive quadrature to marginalize Gaussian process hyperparameters in place of the more common slice sampling approach.
Subjects/Keywords: Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
McLeod, M. (2018). Optimizing Bayesian optimization. (Doctoral Dissertation). University of Oxford. Retrieved from http://ora.ox.ac.uk/objects/uuid:c35f26ba-07ec-4830-ac37-39b37d36a8b3 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.786122
Chicago Manual of Style (16th Edition):
McLeod, Mark. “Optimizing Bayesian optimization.” 2018. Doctoral Dissertation, University of Oxford. Accessed March 04, 2021.
http://ora.ox.ac.uk/objects/uuid:c35f26ba-07ec-4830-ac37-39b37d36a8b3 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.786122.
MLA Handbook (7th Edition):
McLeod, Mark. “Optimizing Bayesian optimization.” 2018. Web. 04 Mar 2021.
Vancouver:
McLeod M. Optimizing Bayesian optimization. [Internet] [Doctoral dissertation]. University of Oxford; 2018. [cited 2021 Mar 04].
Available from: http://ora.ox.ac.uk/objects/uuid:c35f26ba-07ec-4830-ac37-39b37d36a8b3 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.786122.
Council of Science Editors:
McLeod M. Optimizing Bayesian optimization. [Doctoral Dissertation]. University of Oxford; 2018. Available from: http://ora.ox.ac.uk/objects/uuid:c35f26ba-07ec-4830-ac37-39b37d36a8b3 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.786122
California State Polytechnic University – Pomona
26.
Guo, Mo.
Hair Segmentation using Deep Learning.
Degree: MS, Department of Computer Science, 2020, California State Polytechnic University – Pomona
URL: http://hdl.handle.net/10211.3/214986
► Human appearances are always characterized by hair. Hair segmentation in images is helpful in modern applications, such as face and hair modeling, video surveillance or…
(more)
▼ Human appearances are always characterized by hair. Hair segmentation in images is helpful in modern applications, such as face and hair modeling, video surveillance or even gender recognition. In this project we tackle the problem of hair segmentation from both unconstrained and constrained view by relying on textures, without information on shape of head or location of face, nor using body-part classifiers. For this project we???ll try to implement the segmentation part by going over deep
learning approaches rely on small training datasets of face images captured in the wild. We???ll demonstrate the importance of carefully designed datasets and data augmentation strategies for handling challenging occlusions such as hands. We???ll try to improve the efficiency and accuracy of existing segmentation networks using an architecture based on two-stream deconvolution networks and shared convolution network. Furthermore, the resulting segmentation can be directly used by application based on website, such as virtual make-up or hair color replacement.
Advisors/Committee Members: Ji, Hao (advisor), F. Sang, Daisy (committee member).
Subjects/Keywords: machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Guo, M. (2020). Hair Segmentation using Deep Learning. (Masters Thesis). California State Polytechnic University – Pomona. Retrieved from http://hdl.handle.net/10211.3/214986
Chicago Manual of Style (16th Edition):
Guo, Mo. “Hair Segmentation using Deep Learning.” 2020. Masters Thesis, California State Polytechnic University – Pomona. Accessed March 04, 2021.
http://hdl.handle.net/10211.3/214986.
MLA Handbook (7th Edition):
Guo, Mo. “Hair Segmentation using Deep Learning.” 2020. Web. 04 Mar 2021.
Vancouver:
Guo M. Hair Segmentation using Deep Learning. [Internet] [Masters thesis]. California State Polytechnic University – Pomona; 2020. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/10211.3/214986.
Council of Science Editors:
Guo M. Hair Segmentation using Deep Learning. [Masters Thesis]. California State Polytechnic University – Pomona; 2020. Available from: http://hdl.handle.net/10211.3/214986
Universidad de Cantabria
27.
Ruiz Martínez, Estela.
Machine learning methods for the prediction of non-metallic inclusions in steel wires for tire reinforcement: Métodos machine learning para la predicción de inclusiones no metálicas en alambres de acero para refuerzo de neumáticos.
Degree: Máster en Ciencia de Datos, 2019, Universidad de Cantabria
URL: http://hdl.handle.net/10902/16903
► ABSTRACT: Non-metallic inclusions are unavoidably produced during steel casting resulting in lower mechanical strength and other detrimental effects. This study was aimed at developing a…
(more)
▼ ABSTRACT: Non-metallic inclusions are unavoidably produced during steel casting resulting in lower mechanical strength and other detrimental effects. This study was aimed at developing a reliable
Machine Learning algorithm to classify castings of steel for tire reinforcement depending on the number and properties of inclusions, experimentally determined.
855 observations were available for training, validation and testing the algorithms, obtained from the quality control of the steel. 140 parameters are monitored during fabrication, which are the features of the analysis; the output is 1 or 0 depending on whether the casting is rejected or not.
The following algorithms have been employed: Logistic Regression, K-Nearest Neighbors, Support Vector Classifier (linear and RBF kernels), Random Forests, AdaBoost, Gradient Boosting and Artificial Neural Networks. The reduced value of the rejection rate implies that classification must be carried out on an imbalanced dataset. Resampling methods and specific scores for imbalanced datasets (Recall, Precision and AUC rather than Accuracy) were used. Random Forest was the most successful method providing an AUC in the test set of 0.85. No significant improvements were detected after resampling.
The improvement derived from implementing this algorithm in the sampling procedure for quality control during steelmaking has been quantified. In this sense, it has been proved that this tool allows the samples with a higher probability of being rejected to be selected, thus improving the effectiveness of the quality control. In addition, the optimized Random Forest has enabled to identify the most important features, which have been satisfactorily interpreted on a metallurgical basis.
Advisors/Committee Members: Lloret Iglesias, Lara (advisor), Universidad de Cantabria (other).
Subjects/Keywords: Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Ruiz Martínez, E. (2019). Machine learning methods for the prediction of non-metallic inclusions in steel wires for tire reinforcement: Métodos machine learning para la predicción de inclusiones no metálicas en alambres de acero para refuerzo de neumáticos. (Masters Thesis). Universidad de Cantabria. Retrieved from http://hdl.handle.net/10902/16903
Chicago Manual of Style (16th Edition):
Ruiz Martínez, Estela. “Machine learning methods for the prediction of non-metallic inclusions in steel wires for tire reinforcement: Métodos machine learning para la predicción de inclusiones no metálicas en alambres de acero para refuerzo de neumáticos.” 2019. Masters Thesis, Universidad de Cantabria. Accessed March 04, 2021.
http://hdl.handle.net/10902/16903.
MLA Handbook (7th Edition):
Ruiz Martínez, Estela. “Machine learning methods for the prediction of non-metallic inclusions in steel wires for tire reinforcement: Métodos machine learning para la predicción de inclusiones no metálicas en alambres de acero para refuerzo de neumáticos.” 2019. Web. 04 Mar 2021.
Vancouver:
Ruiz Martínez E. Machine learning methods for the prediction of non-metallic inclusions in steel wires for tire reinforcement: Métodos machine learning para la predicción de inclusiones no metálicas en alambres de acero para refuerzo de neumáticos. [Internet] [Masters thesis]. Universidad de Cantabria; 2019. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/10902/16903.
Council of Science Editors:
Ruiz Martínez E. Machine learning methods for the prediction of non-metallic inclusions in steel wires for tire reinforcement: Métodos machine learning para la predicción de inclusiones no metálicas en alambres de acero para refuerzo de neumáticos. [Masters Thesis]. Universidad de Cantabria; 2019. Available from: http://hdl.handle.net/10902/16903
California State Polytechnic University – Pomona
28.
Bailey, Kevin.
Statistical Learning for Esports Match Prediction.
Degree: MS, Department of Mathematics and Statistics, 2020, California State Polytechnic University – Pomona
URL: http://hdl.handle.net/10211.3/216025
► Traditional sports have been a hot topic for many statistical and machine learning models. Not only for prediction, but these models can help to make…
(more)
▼ Traditional sports have been a hot topic for many statistical and
machine learning models. Not only for prediction, but these models can help to make informed decisions on fantasy drafts as well. When only caring about the result of a match, many popular classifications such as logistic regression and support vector machines (SVMs) are used. However, one statistical
learning concept that has been gaining traction, but is still not widely used, is that of Gaussian processes (GPs). This paper aims to analyze of many different models that are common in practice and give a quick rundown on the theory of them, as well as going into Gaussian processes to see how they will compare. Our application is to apply these models on a League of Legends data set of professional matches, and comparing them to see which one works best for this classification of winning or losing. Theoretically, Gaussian processes should be a great fit for this problem since local observations help the predictions more than distant ones, and the domain of some very important factors (gold and gold-difference) tend to be clustered around the average; in fact, it is nearly impossible to stray extremely far from that average.
Advisors/Committee Members: Risk, Jimmy (advisor), King, Adam (committee member).
Subjects/Keywords: machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Bailey, K. (2020). Statistical Learning for Esports Match Prediction. (Masters Thesis). California State Polytechnic University – Pomona. Retrieved from http://hdl.handle.net/10211.3/216025
Chicago Manual of Style (16th Edition):
Bailey, Kevin. “Statistical Learning for Esports Match Prediction.” 2020. Masters Thesis, California State Polytechnic University – Pomona. Accessed March 04, 2021.
http://hdl.handle.net/10211.3/216025.
MLA Handbook (7th Edition):
Bailey, Kevin. “Statistical Learning for Esports Match Prediction.” 2020. Web. 04 Mar 2021.
Vancouver:
Bailey K. Statistical Learning for Esports Match Prediction. [Internet] [Masters thesis]. California State Polytechnic University – Pomona; 2020. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/10211.3/216025.
Council of Science Editors:
Bailey K. Statistical Learning for Esports Match Prediction. [Masters Thesis]. California State Polytechnic University – Pomona; 2020. Available from: http://hdl.handle.net/10211.3/216025
University of Victoria
29.
Lam, Newman Ming Ki.
Learning in the real world environment: a classification model based on sensitivity to within-dimension and between-category variation of feature frequencies.
Degree: School of Public Administration, 2018, University of Victoria
URL: https://dspace.library.uvic.ca//handle/1828/9498
► Research on machine learning has taken numerous different directions. The present study focussed on the microstructural characteristics of learning systems. It was postulated that learning…
(more)
▼ Research on
machine learning has taken numerous different
directions. The present study focussed on the microstructural
characteristics of
learning systems. It was
postulated that
learning systems consist of a macrostructure
which controls the flow of information, and a
micro-structure which manipulates information for decision
making. A review of the literature suggested that the basic
function of the micro-structure of
learning systems was to
make a choice among a set of alternatives. This decision
function was then equated with the task of making
classification decisions. On the basis of the requirements
for practical
learning systems, the feature frequency
approach was chosen for model development. An analysis of
the feature frequency approach indicated that an effective
model must be sensitive to both within-dimension and
between-category variations in frequencies. A model was
then developed to provide for such sensitivities. The model
was based on the Bayes' Theorem with an assumption of
uniform prior probability of occurrence for the categories.
This model was tested using data collected for
neuropsychological diagnosis of children. Results of the
tests showed that the model was capable of
learning and
provided a satisfactory level of performance. The
performance of the model was compared with that of other
models designed for the same purpose. The other models
included NEXSYS, a rule-based system specially design for
this type of diagnosis, discriminant analysis, which is a
statistical technique widely used for pattern recognition,
and neural networks, which attempt to simulate the neural
activities of the brain. Results of the tests showed that
the model's performance was comparable to that of the other
models. Further analysis indicated that the model has certain advantages in that it has a simple structure, is
capable of explaining its decisions, and is more efficient
than the other models.
Advisors/Committee Members: MacGregor, James (supervisor).
Subjects/Keywords: Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Lam, N. M. K. (2018). Learning in the real world environment: a classification model based on sensitivity to within-dimension and between-category variation of feature frequencies. (Thesis). University of Victoria. Retrieved from https://dspace.library.uvic.ca//handle/1828/9498
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Chicago Manual of Style (16th Edition):
Lam, Newman Ming Ki. “Learning in the real world environment: a classification model based on sensitivity to within-dimension and between-category variation of feature frequencies.” 2018. Thesis, University of Victoria. Accessed March 04, 2021.
https://dspace.library.uvic.ca//handle/1828/9498.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
MLA Handbook (7th Edition):
Lam, Newman Ming Ki. “Learning in the real world environment: a classification model based on sensitivity to within-dimension and between-category variation of feature frequencies.” 2018. Web. 04 Mar 2021.
Vancouver:
Lam NMK. Learning in the real world environment: a classification model based on sensitivity to within-dimension and between-category variation of feature frequencies. [Internet] [Thesis]. University of Victoria; 2018. [cited 2021 Mar 04].
Available from: https://dspace.library.uvic.ca//handle/1828/9498.
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
Council of Science Editors:
Lam NMK. Learning in the real world environment: a classification model based on sensitivity to within-dimension and between-category variation of feature frequencies. [Thesis]. University of Victoria; 2018. Available from: https://dspace.library.uvic.ca//handle/1828/9498
Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation
30.
Fan, Shuangfei.
Deep Representation Learning on Labeled Graphs.
Degree: PhD, Computer Science and Applications, 2020, Virginia Tech
URL: http://hdl.handle.net/10919/96596
► Graphs are one of the most important and powerful data structures for conveying the complex and correlated information among data points. In this research, we…
(more)
▼ Graphs are one of the most important and powerful data structures for conveying the complex and correlated information among data points. In this research, we aim to provide more robust and accurate models for some graph specific tasks, such as collective classification and graph generation, by designing deep
learning models to learn better task-specific representations for graphs. First, we studied the collective classification problem in graphs and proposed recurrent collective classification, a variant of the iterative classification algorithm that is more robust to situations where predictions are noisy or inaccurate. Then we studied the problem of graph generation using deep generative models. We first proposed a deep generative model using the GAN framework that generates labeled graphs. Then in order to support more applications and also get more control over the generated graphs, we extended the problem of graph generation to conditional graph generation which can then be applied to various applications for modeling graph evolution and transformation.
Advisors/Committee Members: Huang, Bert (committeechair), Neville, Jennifer (committee member), Abbott, Amos L. (committee member), Ramakrishnan, Naren (committee member), Reddy, Chandan K. (committee member).
Subjects/Keywords: Machine learning
Record Details
Similar Records
Cite
Share »
Record Details
Similar Records
Cite
« Share
❌
APA ·
Chicago ·
MLA ·
Vancouver ·
CSE |
Export
to Zotero / EndNote / Reference
Manager
APA (6th Edition):
Fan, S. (2020). Deep Representation Learning on Labeled Graphs. (Doctoral Dissertation). Virginia Tech. Retrieved from http://hdl.handle.net/10919/96596
Chicago Manual of Style (16th Edition):
Fan, Shuangfei. “Deep Representation Learning on Labeled Graphs.” 2020. Doctoral Dissertation, Virginia Tech. Accessed March 04, 2021.
http://hdl.handle.net/10919/96596.
MLA Handbook (7th Edition):
Fan, Shuangfei. “Deep Representation Learning on Labeled Graphs.” 2020. Web. 04 Mar 2021.
Vancouver:
Fan S. Deep Representation Learning on Labeled Graphs. [Internet] [Doctoral dissertation]. Virginia Tech; 2020. [cited 2021 Mar 04].
Available from: http://hdl.handle.net/10919/96596.
Council of Science Editors:
Fan S. Deep Representation Learning on Labeled Graphs. [Doctoral Dissertation]. Virginia Tech; 2020. Available from: http://hdl.handle.net/10919/96596
◁ [1] [2] [3] [4] [5] … [377] ▶
. 
Open Access Theses and Dissertations
