subject:(multi task learning)

Arizona State University

1. Zhou, Jiayu. Multi-Task Learning and Its Applications to Biomedical Informatics.

Degree: PhD, Computer Science, 2014, Arizona State University

URL: http://repository.asu.edu/items/25176

► In many fields one needs to build predictive models for a set of related machine learning tasks, such as information retrieval, computer vision and biomedical… (more)

Subjects/Keywords: Computer science; multi-task learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Zhou, J. (2014). Multi-Task Learning and Its Applications to Biomedical Informatics. (Doctoral Dissertation). Arizona State University. Retrieved from http://repository.asu.edu/items/25176

Chicago Manual of Style (16^th Edition):

Zhou, Jiayu. “Multi-Task Learning and Its Applications to Biomedical Informatics.” 2014. Doctoral Dissertation, Arizona State University. Accessed March 09, 2021. http://repository.asu.edu/items/25176.

MLA Handbook (7^th Edition):

Zhou, Jiayu. “Multi-Task Learning and Its Applications to Biomedical Informatics.” 2014. Web. 09 Mar 2021.

Vancouver:

Zhou J. Multi-Task Learning and Its Applications to Biomedical Informatics. [Internet] [Doctoral dissertation]. Arizona State University; 2014. [cited 2021 Mar 09]. Available from: http://repository.asu.edu/items/25176.

Council of Science Editors:

Zhou J. Multi-Task Learning and Its Applications to Biomedical Informatics. [Doctoral Dissertation]. Arizona State University; 2014. Available from: http://repository.asu.edu/items/25176

Tampere University

2. Senhaji, Ali. Incremental Multi-Domain Learning with Domain-Specific Early Exits .

Degree: 2020, Tampere University

URL: https://trepo.tuni.fi/handle/10024/121513

► Deep learning architectures can achieve state-of-the-art results in several computer vision tasks. However, these methods are highly specialized, i.e., for every task from a new… (more)

▼ Deep learning architectures can achieve state-of-the-art results in several computer vision tasks. However, these methods are highly specialized, i.e., for every task from a new domain, an independent new model is required. Multi-domain learning investigates new ways of developing one model capable of solving different tasks from different domains. Most of the multi-domain methods try to maximize the parameter sharing, i.e., domain agnostic parameters or base model, and minimize domain-specific parameters. In this archetype, all of the target domains utilize all of the base network. Given that domains come in different levels of difficulty, this leads to inefficient use of the base model to solve tasks in easier domains. In this thesis, we examine the adaptive use of the base model parameters. We propose a novel adaptive approach for incremental multi-domain learning, where different parts of the base network are adapted depending on the level of complexity of each individual domain. The aim is to reach an efficient use of the base network while maintaining a high performance. Developing efficient models with the most optimal capacity is important for a multitude of applications. The proposed adaptive method achieves comparable performance to adapting the whole base network for easier domains while reducing by far the number of parameters. This leads to efficient multi-domain learning solutions and can be useful in many applications (e.g., budget inference within edge devices). We investigated the use of the proposed approach for residual networks. High performance comparable to using the whole network was achieved for domains with easy and intermediate levels of difficulty, with only 4.7% and 23.8% of the parameters, respectively. We have used a benchmark of ten visually different datasets, to solve problems including recognizing handwritten characters, classifying flowers, detecting pedestrians, classifying aircraft, and detecting actions from real-life video snapshots. Our adaptive method achieved a mean accuracy of 72.79%, using only 15% of the parameters required to have ten different fine-tuned networks, compared to 73.44% mean accuracy. Thus, our results confirm our hypothesis that it is not necessary to use the whole base network for all the domains, and easier domains can be more efficiently parameterized with the proposed method.

Subjects/Keywords: multi-domain learning ; early exits ; domain adaptation ; multi-task learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Senhaji, A. (2020). Incremental Multi-Domain Learning with Domain-Specific Early Exits . (Masters Thesis). Tampere University. Retrieved from https://trepo.tuni.fi/handle/10024/121513

Chicago Manual of Style (16^th Edition):

Senhaji, Ali. “Incremental Multi-Domain Learning with Domain-Specific Early Exits .” 2020. Masters Thesis, Tampere University. Accessed March 09, 2021. https://trepo.tuni.fi/handle/10024/121513.

MLA Handbook (7^th Edition):

Senhaji, Ali. “Incremental Multi-Domain Learning with Domain-Specific Early Exits .” 2020. Web. 09 Mar 2021.

Vancouver:

Senhaji A. Incremental Multi-Domain Learning with Domain-Specific Early Exits . [Internet] [Masters thesis]. Tampere University; 2020. [cited 2021 Mar 09]. Available from: https://trepo.tuni.fi/handle/10024/121513.

Council of Science Editors:

Senhaji A. Incremental Multi-Domain Learning with Domain-Specific Early Exits . [Masters Thesis]. Tampere University; 2020. Available from: https://trepo.tuni.fi/handle/10024/121513

University of Minnesota

3. Karpatne, Anuj. Predictive Learning with Heterogeneity in Populations.

Degree: PhD, Computer Science, 2017, University of Minnesota

URL: http://hdl.handle.net/11299/192667

► Predictive learning forms the backbone of several data-driven systems powering scientific as well as commercial applications, e.g., filtering spam messages, detecting faces in images, forecasting… (more)

▼ Predictive learning forms the backbone of several data-driven systems powering scientific as well as commercial applications, e.g., filtering spam messages, detecting faces in images, forecasting health risks, and mapping ecological resources. However, one of the major challenges in applying standard predictive learning methods in real-world applications is the heterogeneity in populations of data instances, i.e., different groups (or populations) of data instances show different nature of predictive relationships. For example, different populations of human subjects may show different risks for a disease even if they have similar diagnosis reports, depending on their ethnic profiles, medical history, and lifestyle choices. In the presence of population heterogeneity, a central challenge is that the training data comprises of instances belonging from multiple populations, and the instances in the test set may be from a different population than that of the training instances. This limits the effectiveness of standard predictive learning frameworks that are based on the assumption that the instances are independent and identically distributed (i.i.d), which are ideally true only in simplistic settings. This thesis introduces several ways of learning predictive models with heterogeneity in populations, by incorporating information about the context of every data instance, which is available in varying types and formats in different application settings. It introduces a novel multi-task learning framework for problems where we have access to some ancillary variables that can be grouped to produce homogeneous partitions of data instances, thus addressing the heterogeneity in populations. This thesis also introduces a novel strategy for constructing mode-specific ensembles in binary classification settings, where each class shows multi-modal distribution due to the heterogeneity in their populations. When the context of data instances is implicitly defined such that the test data is known to comprise of contextually similar groups, this thesis presents a novel framework for adapting classification decisions using the group-level properties of test instances. This thesis also builds the foundations of a novel paradigm of scientific discovery, termed as theory-guided data science, that seeks to explore the full potential of data science methods but without ignoring the treasure of knowledge contained in scientific theories and principles.

Subjects/Keywords: data mining; ensemble learning; machine learning; multi-modality; multi-task learning; population heterogeneity

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Karpatne, A. (2017). Predictive Learning with Heterogeneity in Populations. (Doctoral Dissertation). University of Minnesota. Retrieved from http://hdl.handle.net/11299/192667

Chicago Manual of Style (16^th Edition):

Karpatne, Anuj. “Predictive Learning with Heterogeneity in Populations.” 2017. Doctoral Dissertation, University of Minnesota. Accessed March 09, 2021. http://hdl.handle.net/11299/192667.

MLA Handbook (7^th Edition):

Karpatne, Anuj. “Predictive Learning with Heterogeneity in Populations.” 2017. Web. 09 Mar 2021.

Vancouver:

Karpatne A. Predictive Learning with Heterogeneity in Populations. [Internet] [Doctoral dissertation]. University of Minnesota; 2017. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/11299/192667.

Council of Science Editors:

Karpatne A. Predictive Learning with Heterogeneity in Populations. [Doctoral Dissertation]. University of Minnesota; 2017. Available from: http://hdl.handle.net/11299/192667

AUT University

4. Fan, Liu. Minimum Enclosing Ball-based Learner Independent Knowledge Transfer for Correlated Multi-task Learning .

Degree: 2011, AUT University

URL: http://hdl.handle.net/10292/1120

► Multi-Task Learning (MTL), as opposed to Single Task Learning (STL), has become a hot topic in machine learning research. For many real world problems in… (more)

Subjects/Keywords: Multi-task Learning; Knowledge Transfer; Correlated multi-task learning; Minimum Enclosing Ball; Machine Learning; Knowledge Sharing; Learner Independence

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Fan, L. (2011). Minimum Enclosing Ball-based Learner Independent Knowledge Transfer for Correlated Multi-task Learning . (Thesis). AUT University. Retrieved from http://hdl.handle.net/10292/1120

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16^th Edition):

Fan, Liu. “Minimum Enclosing Ball-based Learner Independent Knowledge Transfer for Correlated Multi-task Learning .” 2011. Thesis, AUT University. Accessed March 09, 2021. http://hdl.handle.net/10292/1120.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7^th Edition):

Fan, Liu. “Minimum Enclosing Ball-based Learner Independent Knowledge Transfer for Correlated Multi-task Learning .” 2011. Web. 09 Mar 2021.

Vancouver:

Fan L. Minimum Enclosing Ball-based Learner Independent Knowledge Transfer for Correlated Multi-task Learning . [Internet] [Thesis]. AUT University; 2011. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/10292/1120.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Fan L. Minimum Enclosing Ball-based Learner Independent Knowledge Transfer for Correlated Multi-task Learning . [Thesis]. AUT University; 2011. Available from: http://hdl.handle.net/10292/1120

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

5. Moura, Simon. Apprentissage multi-cibles : théorie et applications : Multi-output learning : theory and applications.

Degree: Docteur es, Informatique, 2018, Université Grenoble Alpes (ComUE)

URL: http://www.theses.fr/2018GREAM085

►

Cette thèse traite du problème de l'apprentissage automatique supervisé dans le cas ou l'on considère plusieurs sorties, potentiellement de différent types. Nous proposons d'explorer trois… (more)

▼

Cette thèse traite du problème de l'apprentissage automatique supervisé dans le cas ou l'on considère plusieurs sorties, potentiellement de différent types. Nous proposons d'explorer trois différents axes de recherche en rapport avec ce sujet. Dans un premier temps, nous nous concentrons sur le cas homogène et proposons un cadre théorique pour étudier la consistance des problèmes multi-labels dans le cas de l'utilisation de chaîne de classifieurs. Ensuite, en nous plaçant dans ce cadre, nous proposons une borne de Rademacher sur l'erreur de généralisation pour tous les classifieurs de la chaîne et exposons deux facteurs de dépendance reliant les sorties les unes aux autres. Dans un deuxième temps, nous développons et analysons la performance de modèles en lien avec la théorie proposée. Toujours dans le cadre de l'apprentissage avec plusieurs sorties homogènes, nous proposons un modèle basé sur des réseaux de neurones pour l'analyse de sentiments à grain fin. Enfin, nous proposons un cadre et une étude empirique qui montrent la pertinence de l'apprentissage multi-objectif dans le cas de multiples sorties hétérogènes.

In this thesis, we study the problem of learning with multiple outputs related to different tasks, such as classification and ranking. In this line of research, we explored three different axes. First we proposed a theoretical framework that can be used to show the consistency of multi-label learning in the case of classifier chains, where outputs are homogeneous. Based on this framework, we proposed Rademacher generalization error bound made by any classifier in the chain and exhibit dependency factors relating each output to the others. As a result, we introduced multiple strategies to learn classifier chains and select an order for the chain. Still focusing on the homogeneous multi-output framework, we proposed a neural network based solution for fine-grained sentiment analysis and show the efficiency of the approach. Finally, we proposed a framework and an empirical study showing the interest of learning with multiple tasks, even when the outputs are of different types.

Advisors/Committee Members: Amini, Massih-Reza (thesis director).

Subjects/Keywords: Apprentissage multi-Cibles; Apprentissage statistique; Apprentissage multi-Label; Multi-Output learning; Multi-Task learning; Statisticial learning; 004

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Moura, S. (2018). Apprentissage multi-cibles : théorie et applications : Multi-output learning : theory and applications. (Doctoral Dissertation). Université Grenoble Alpes (ComUE). Retrieved from http://www.theses.fr/2018GREAM085

Chicago Manual of Style (16^th Edition):

Moura, Simon. “Apprentissage multi-cibles : théorie et applications : Multi-output learning : theory and applications.” 2018. Doctoral Dissertation, Université Grenoble Alpes (ComUE). Accessed March 09, 2021. http://www.theses.fr/2018GREAM085.

MLA Handbook (7^th Edition):

Moura, Simon. “Apprentissage multi-cibles : théorie et applications : Multi-output learning : theory and applications.” 2018. Web. 09 Mar 2021.

Vancouver:

Moura S. Apprentissage multi-cibles : théorie et applications : Multi-output learning : theory and applications. [Internet] [Doctoral dissertation]. Université Grenoble Alpes (ComUE); 2018. [cited 2021 Mar 09]. Available from: http://www.theses.fr/2018GREAM085.

Council of Science Editors:

Moura S. Apprentissage multi-cibles : théorie et applications : Multi-output learning : theory and applications. [Doctoral Dissertation]. Université Grenoble Alpes (ComUE); 2018. Available from: http://www.theses.fr/2018GREAM085

6. Faddoul, Jean-Baptiste. Méthodes d’ensembles pour l’apprentissage multi-tâche avec des tâches hétérogènes et sans restrictions : Ensemble Methods to Learn Multiple Heterogenous Tasks without Restrictions.

Degree: Docteur es, Informatique, 2012, Lille 3

URL: http://www.theses.fr/2012LIL30059

►

Apprendre des tâches simultanément peut améliorer la performance de prédiction par rapport à l'apprentissage de ces tâches de manière indépendante. Dans cette thèse, nous considérons… (more)

▼

Apprendre des tâches simultanément peut améliorer la performance de prédiction par rapport à l'apprentissage de ces tâches de manière indépendante. Dans cette thèse, nous considérons l'apprentissage multi-tâche lorsque le nombre de tâches est grand. En outre, nous débattons des restrictions imposées sur les tâches. Ces restrictions peuvent être trouvées dans les méthodes de l'état de l'art. Plus précisément on trouve les restrictions suivantes : l'imposition du même espace d'étiquette sur les tâches, l'exigence des mêmes exemples d'apprentissage entre tâches et / ou supposant une hypothèse de corrélation globale entre tâches. Nous proposons des nouveaux classificateurs multi-tâches qui relaxent les restrictions précédentes. Nos classificateurs sont considérés en fonction de la théorie de l'apprentissage PAC des classifieurs faibles, donc, afin de parvenir à un faible taux d'erreur de classification, un ensemble de ces classifieurs faibles doivent être appris. Ce cadre est appelé l'apprentissage d'ensembles, dans lequel nous proposons un algorithme d'apprentissage multi-tâche inspiré de l'algorithme Adaboost pour seule tâche. Différentes variantes sont proposées également, à savoir, les forêts aléatoires pour le multi-tâche, c'est une méthode d'apprentissage d'ensemble, mais fondée sur le principe statistique d'échantillonnage Bootstrap. Enfin, nous donnons une validation expérimentale qui montre que l'approche sur-performe des méthodes existantes et permet d'apprendre des nouvelles configurations de tâches qui ne correspondent pas aux méthodes de l'état de l'art.

Learning multiple related tasks jointly by exploiting their underlying shared knowledge can improve the predictive performance on every task compared to learning them individually. In this thesis, we address the problem of multi-task learning (MTL) when the tasks are heterogenous: they do not share the same labels (eventually with different number of labels), they do not require shared examples. In addition, no prior assumption about the relatedness pattern between tasks is made. Our contribution to multi-task learning lies in the framework of en- semble learning where the learned function consists normally of an ensemble of "weak " hypothesis aggregated together by an ensemble learning algorithm (Boosting, Bagging, etc.). We propose two approaches to cope with heterogenous tasks without making prior assumptions about the relatedness patterns. For each approach, we devise novel multi-task weak hypothesis along with their learning algorithms then we adapt a boosting algorithm to the multi-task setting. In the first approach, the weak classi ers we consider are 2-level decision stumps for di erent tasks. A weak classi er assigns a class to each instance on two tasks and abstain on other tasks. The weak classi ers allow to handle dependencies between tasks on the instance space. We introduce di fferent effi cient weak learners. We then consider Adaboost with weak classi ers which can abstain and adapt it to multi-task learning. In an empirical study, we…

Advisors/Committee Members: Gilleron, Rémi (thesis director).

Subjects/Keywords: Apprentissage automatique; Boosting (algorithmes); Fonctionnement multitâche; Machine Learning; Multi-Task Learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Faddoul, J. (2012). Méthodes d’ensembles pour l’apprentissage multi-tâche avec des tâches hétérogènes et sans restrictions : Ensemble Methods to Learn Multiple Heterogenous Tasks without Restrictions. (Doctoral Dissertation). Lille 3. Retrieved from http://www.theses.fr/2012LIL30059

Chicago Manual of Style (16^th Edition):

Faddoul, Jean-Baptiste. “Méthodes d’ensembles pour l’apprentissage multi-tâche avec des tâches hétérogènes et sans restrictions : Ensemble Methods to Learn Multiple Heterogenous Tasks without Restrictions.” 2012. Doctoral Dissertation, Lille 3. Accessed March 09, 2021. http://www.theses.fr/2012LIL30059.

MLA Handbook (7^th Edition):

Faddoul, Jean-Baptiste. “Méthodes d’ensembles pour l’apprentissage multi-tâche avec des tâches hétérogènes et sans restrictions : Ensemble Methods to Learn Multiple Heterogenous Tasks without Restrictions.” 2012. Web. 09 Mar 2021.

Vancouver:

Faddoul J. Méthodes d’ensembles pour l’apprentissage multi-tâche avec des tâches hétérogènes et sans restrictions : Ensemble Methods to Learn Multiple Heterogenous Tasks without Restrictions. [Internet] [Doctoral dissertation]. Lille 3; 2012. [cited 2021 Mar 09]. Available from: http://www.theses.fr/2012LIL30059.

Council of Science Editors:

Faddoul J. Méthodes d’ensembles pour l’apprentissage multi-tâche avec des tâches hétérogènes et sans restrictions : Ensemble Methods to Learn Multiple Heterogenous Tasks without Restrictions. [Doctoral Dissertation]. Lille 3; 2012. Available from: http://www.theses.fr/2012LIL30059

University of Cambridge

7. Bruinsma, Wessel. The Generalised Gaussian Process Convolution Model.

Degree: MPhil, 2016, University of Cambridge

URL: https://www.repository.cam.ac.uk/handle/1810/273357

► This thesis formulates the Generalised Gaussian Process Convolution Model (GGPCM), which is a generalisation of the Gaussian Process Convolution Model presented by Tobar et al.… (more)

Subjects/Keywords: machine learning; Gaussian process; kernel; nonparametric kernel; multi-task learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Bruinsma, W. (2016). The Generalised Gaussian Process Convolution Model. (Masters Thesis). University of Cambridge. Retrieved from https://www.repository.cam.ac.uk/handle/1810/273357

Chicago Manual of Style (16^th Edition):

Bruinsma, Wessel. “The Generalised Gaussian Process Convolution Model.” 2016. Masters Thesis, University of Cambridge. Accessed March 09, 2021. https://www.repository.cam.ac.uk/handle/1810/273357.

MLA Handbook (7^th Edition):

Bruinsma, Wessel. “The Generalised Gaussian Process Convolution Model.” 2016. Web. 09 Mar 2021.

Vancouver:

Bruinsma W. The Generalised Gaussian Process Convolution Model. [Internet] [Masters thesis]. University of Cambridge; 2016. [cited 2021 Mar 09]. Available from: https://www.repository.cam.ac.uk/handle/1810/273357.

Council of Science Editors:

Bruinsma W. The Generalised Gaussian Process Convolution Model. [Masters Thesis]. University of Cambridge; 2016. Available from: https://www.repository.cam.ac.uk/handle/1810/273357

Delft University of Technology

8. Kaniouras, Pantelis (author). Road Detection from Remote Sensing Imagery.

Degree: 2020, Delft University of Technology

URL: http://resolver.tudelft.nl/uuid:21fc20a8-455d-4583-9698-4fea04516f03

►

Road network maps facilitate a great number of applications in our everyday life. However, their automatic creation is a difficult task, and so far, published… (more)

Subjects/Keywords: multi-task learning; deep learning; road detection; Convolutional Neural Network

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Kaniouras, P. (. (2020). Road Detection from Remote Sensing Imagery. (Masters Thesis). Delft University of Technology. Retrieved from http://resolver.tudelft.nl/uuid:21fc20a8-455d-4583-9698-4fea04516f03

Chicago Manual of Style (16^th Edition):

Kaniouras, Pantelis (author). “Road Detection from Remote Sensing Imagery.” 2020. Masters Thesis, Delft University of Technology. Accessed March 09, 2021. http://resolver.tudelft.nl/uuid:21fc20a8-455d-4583-9698-4fea04516f03.

MLA Handbook (7^th Edition):

Kaniouras, Pantelis (author). “Road Detection from Remote Sensing Imagery.” 2020. Web. 09 Mar 2021.

Vancouver:

Kaniouras P(. Road Detection from Remote Sensing Imagery. [Internet] [Masters thesis]. Delft University of Technology; 2020. [cited 2021 Mar 09]. Available from: http://resolver.tudelft.nl/uuid:21fc20a8-455d-4583-9698-4fea04516f03.

Council of Science Editors:

Kaniouras P(. Road Detection from Remote Sensing Imagery. [Masters Thesis]. Delft University of Technology; 2020. Available from: http://resolver.tudelft.nl/uuid:21fc20a8-455d-4583-9698-4fea04516f03

Tampere University

9. Khan, Amna. Comparison of machine learning approaches for classification of invoices .

Degree: 2020, Tampere University

URL: https://trepo.tuni.fi/handle/10024/120493

► Machine learning has become one of the leading sciences governing modern world. Various disciplines specifically neural networks have recently gained a lot of attention due… (more)

▼ Machine learning has become one of the leading sciences governing modern world. Various disciplines specifically neural networks have recently gained a lot of attention due to its widespread applications. With the recent advances in the technology the resulting big data has augmented the need of bigger means of storage, analysis and henceforth utilization. This not only implies the efficient use of available techniques but suggests surge in the development of new algorithms and techniques. In this project, three different machine learning approaches were implemented utilizing the open source library of keras on TensorFlow as a proof of concept for the task of intelligent invoice automation. The performance of these approaches for improved business on data of invoices has been analysed using the data of two customers with two target attributes per customer as a dataset. The behaviour of neural network hyper-parameters using matplotlib and TensorBoard was empirically calculated and investigated. As part of the first approach, the standard way of implementing predictive algorithm using neural network was followed. Moreover, the hyper-parameters search space was fine-tuned, and the resulting model was studied by grid search on those hyper-parameters. This strategy of hyper-parameters was followed in the next two approaches as well. In the second approach, not only further possible improvement in prediction accuracy is achieved but also the dependency between the two target attributes by using multi-task learning was determined. As per the third implemented approach, the use of continual learning on invoices for postings was analysed. This investigation, that involves the comparison of varied machine learning approaches has broad significance in approving the currently available algorithms for handling such data and suggests means for improvement as well. It holds great prospects, including but not limited to future implementation of such approaches in the domain of finance towards improved customer experience, fraud detection and ease in the assessments of assets etc.

Subjects/Keywords: Machine Learning ; Invoice prediction ; Neural Networks ; Multi-task learning ; Continual Learning ; Deep Learning in Finance

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Khan, A. (2020). Comparison of machine learning approaches for classification of invoices . (Masters Thesis). Tampere University. Retrieved from https://trepo.tuni.fi/handle/10024/120493

Chicago Manual of Style (16^th Edition):

Khan, Amna. “Comparison of machine learning approaches for classification of invoices .” 2020. Masters Thesis, Tampere University. Accessed March 09, 2021. https://trepo.tuni.fi/handle/10024/120493.

MLA Handbook (7^th Edition):

Khan, Amna. “Comparison of machine learning approaches for classification of invoices .” 2020. Web. 09 Mar 2021.

Vancouver:

Khan A. Comparison of machine learning approaches for classification of invoices . [Internet] [Masters thesis]. Tampere University; 2020. [cited 2021 Mar 09]. Available from: https://trepo.tuni.fi/handle/10024/120493.

Council of Science Editors:

Khan A. Comparison of machine learning approaches for classification of invoices . [Masters Thesis]. Tampere University; 2020. Available from: https://trepo.tuni.fi/handle/10024/120493

University of Kansas

10. Li, Xiaoli. Constructivism Learning: A Learning Paradigm for Transparent Predictive Analytics.

Degree: PhD, Electrical Engineering & Computer Science, 2018, University of Kansas

URL: http://hdl.handle.net/1808/27594

► Aiming to achieve the learning capabilities possessed by intelligent beings, especially human, researchers in machine learning field have the long-standing tradition of bor- rowing ideas… (more)

Subjects/Keywords: Computer science; Bayesian Nonparametrics; Constructivism Learning; Multi-task Learning; Transparent Machine Learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Li, X. (2018). Constructivism Learning: A Learning Paradigm for Transparent Predictive Analytics. (Doctoral Dissertation). University of Kansas. Retrieved from http://hdl.handle.net/1808/27594

Chicago Manual of Style (16^th Edition):

Li, Xiaoli. “Constructivism Learning: A Learning Paradigm for Transparent Predictive Analytics.” 2018. Doctoral Dissertation, University of Kansas. Accessed March 09, 2021. http://hdl.handle.net/1808/27594.

MLA Handbook (7^th Edition):

Li, Xiaoli. “Constructivism Learning: A Learning Paradigm for Transparent Predictive Analytics.” 2018. Web. 09 Mar 2021.

Vancouver:

Li X. Constructivism Learning: A Learning Paradigm for Transparent Predictive Analytics. [Internet] [Doctoral dissertation]. University of Kansas; 2018. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/1808/27594.

Council of Science Editors:

Li X. Constructivism Learning: A Learning Paradigm for Transparent Predictive Analytics. [Doctoral Dissertation]. University of Kansas; 2018. Available from: http://hdl.handle.net/1808/27594

Rochester Institute of Technology

11. Sankaran, Prashant. Design and Simulation Analysis of Deep Learning Based Approaches and Multi-Attribute Algorithms for Warehouse Task Selection.

Degree: MS, Industrial and Systems Engineering, 2020, Rochester Institute of Technology

URL: https://scholarworks.rit.edu/theses/10354

► With the growth and adoption of global supply chains and internet technologies, warehouse operations have become more demanding. Particularly, the number of orders being… (more)

Subjects/Keywords: Algorithm; Deep learning; Dynamic environment; Multi attribute task assignment; Simulation; Warehouse

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Sankaran, P. (2020). Design and Simulation Analysis of Deep Learning Based Approaches and Multi-Attribute Algorithms for Warehouse Task Selection. (Masters Thesis). Rochester Institute of Technology. Retrieved from https://scholarworks.rit.edu/theses/10354

Chicago Manual of Style (16^th Edition):

Sankaran, Prashant. “Design and Simulation Analysis of Deep Learning Based Approaches and Multi-Attribute Algorithms for Warehouse Task Selection.” 2020. Masters Thesis, Rochester Institute of Technology. Accessed March 09, 2021. https://scholarworks.rit.edu/theses/10354.

MLA Handbook (7^th Edition):

Sankaran, Prashant. “Design and Simulation Analysis of Deep Learning Based Approaches and Multi-Attribute Algorithms for Warehouse Task Selection.” 2020. Web. 09 Mar 2021.

Vancouver:

Sankaran P. Design and Simulation Analysis of Deep Learning Based Approaches and Multi-Attribute Algorithms for Warehouse Task Selection. [Internet] [Masters thesis]. Rochester Institute of Technology; 2020. [cited 2021 Mar 09]. Available from: https://scholarworks.rit.edu/theses/10354.

Council of Science Editors:

Sankaran P. Design and Simulation Analysis of Deep Learning Based Approaches and Multi-Attribute Algorithms for Warehouse Task Selection. [Masters Thesis]. Rochester Institute of Technology; 2020. Available from: https://scholarworks.rit.edu/theses/10354

University of Illinois – Urbana-Champaign

12. Dave, Mihika. Multimodal machine translation.

Degree: MS, Computer Science, 2018, University of Illinois – Urbana-Champaign

URL: http://hdl.handle.net/2142/101374

► Over the past few years, there has been a lot of progress being made in machine translation through deep learning networks. But there is relatively… (more)

Subjects/Keywords: multimodal machine translation; neural machine translation; multi-task learning; image captioning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Dave, M. (2018). Multimodal machine translation. (Thesis). University of Illinois – Urbana-Champaign. Retrieved from http://hdl.handle.net/2142/101374

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16^th Edition):

Dave, Mihika. “Multimodal machine translation.” 2018. Thesis, University of Illinois – Urbana-Champaign. Accessed March 09, 2021. http://hdl.handle.net/2142/101374.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7^th Edition):

Dave, Mihika. “Multimodal machine translation.” 2018. Web. 09 Mar 2021.

Vancouver:

Dave M. Multimodal machine translation. [Internet] [Thesis]. University of Illinois – Urbana-Champaign; 2018. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/2142/101374.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Dave M. Multimodal machine translation. [Thesis]. University of Illinois – Urbana-Champaign; 2018. Available from: http://hdl.handle.net/2142/101374

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

George Mason University

13. Naik, Azad. Using Multi-Task Learning For Large-Scale Document Classification .

Degree: 2013, George Mason University

URL: http://hdl.handle.net/1920/8479

► Multi-Task Learning (MTL) involves learning of multiple tasks, jointly. It seeks to improve the generalization performance of each task by leveraging the relationships among the… (more)

Subjects/Keywords: Multi-Task Learning; classification; model selection; logistic regression; random projection (hashing)

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Naik, A. (2013). Using Multi-Task Learning For Large-Scale Document Classification . (Thesis). George Mason University. Retrieved from http://hdl.handle.net/1920/8479

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16^th Edition):

Naik, Azad. “Using Multi-Task Learning For Large-Scale Document Classification .” 2013. Thesis, George Mason University. Accessed March 09, 2021. http://hdl.handle.net/1920/8479.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7^th Edition):

Naik, Azad. “Using Multi-Task Learning For Large-Scale Document Classification .” 2013. Web. 09 Mar 2021.

Vancouver:

Naik A. Using Multi-Task Learning For Large-Scale Document Classification . [Internet] [Thesis]. George Mason University; 2013. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/1920/8479.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Naik A. Using Multi-Task Learning For Large-Scale Document Classification . [Thesis]. George Mason University; 2013. Available from: http://hdl.handle.net/1920/8479

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Delft University of Technology

14. Rentroia Pacheco, Barbara (author). Multi-task learning of transcriptomic signatures underlying cancer gene dependencies.

Degree: 2019, Delft University of Technology

URL: http://resolver.tudelft.nl/uuid:2e793ece-4572-4bb6-83e3-541be467cb4f

►

Due to their altered genetic context, cancer cells can become dependent on specific genes for their survival. Such cancer-specific dependencies may represent promising therapeutic targets.… (more)

▼

Due to their altered genetic context, cancer cells can become dependent on specific genes for their survival. Such cancer-specific dependencies may represent promising therapeutic targets. However, knowledge on which molecular features of cancer cells induce specific dependencies is still limited and hampers the development of effective targeted therapies. Several large scale studies have systematically measured the dependency of hundreds of known cancer cell lines on thousands of genes using gene silencing. These data have enabled the learning of supervised models to predict dependencies of cancer cells on each gene based on molecular features of the cells. In particular, linear regression with regularization, such as Elastic Net, has been used to select molecular features associated with such dependencies. Since these approaches model dependencies for each gene independently, the selected features provide limited insight into common mechanisms underlying gene dependency. Moreover, they may fail to identify robust associations with gene dependency due to the small size of the available training data. In this work, we apply a multi-task learning approach (Macau) to learn the relationship between transcriptome and gene dependency in cancer cell lines for multiple genes simultaneously. To do so, Macau projects genes, cancer cell lines and their features into a shared latent space. We explore this latent space to go beyond linking individual transcriptomic features with dependencies, and further associate pathway changes with functionally related genes without enforcing prior knowledge on pathway structure. Although Macau and Elastic Net yield similar predictive performance, they find different kinds of associations. First, Macau favors features that are relevant for predicting dependency across multiple genes. Second, Macau captures inherent functional relationships between genes and leverages these to predict cancer gene dependencies. Additionally, Macau can recover similarities between cancer cell lines belonging to the same cancer type based on their dependencies only. In summary, modelling cancer dependencies simultaneously for multiple genes can reveal underlying mechanisms shared by functionally related genes, which would be missed when learning models independently per gene.

Computer Science | Data Science and Technology

Advisors/Committee Members: de Pinho Gonçalves, Joana (mentor), Delft University of Technology (degree granting institution).

Subjects/Keywords: bioinformatics; multi-task learning; cancer dependencies; cancer genomics

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Rentroia Pacheco, B. (. (2019). Multi-task learning of transcriptomic signatures underlying cancer gene dependencies. (Masters Thesis). Delft University of Technology. Retrieved from http://resolver.tudelft.nl/uuid:2e793ece-4572-4bb6-83e3-541be467cb4f

Chicago Manual of Style (16^th Edition):

Rentroia Pacheco, Barbara (author). “Multi-task learning of transcriptomic signatures underlying cancer gene dependencies.” 2019. Masters Thesis, Delft University of Technology. Accessed March 09, 2021. http://resolver.tudelft.nl/uuid:2e793ece-4572-4bb6-83e3-541be467cb4f.

MLA Handbook (7^th Edition):

Rentroia Pacheco, Barbara (author). “Multi-task learning of transcriptomic signatures underlying cancer gene dependencies.” 2019. Web. 09 Mar 2021.

Vancouver:

Rentroia Pacheco B(. Multi-task learning of transcriptomic signatures underlying cancer gene dependencies. [Internet] [Masters thesis]. Delft University of Technology; 2019. [cited 2021 Mar 09]. Available from: http://resolver.tudelft.nl/uuid:2e793ece-4572-4bb6-83e3-541be467cb4f.

Council of Science Editors:

Rentroia Pacheco B(. Multi-task learning of transcriptomic signatures underlying cancer gene dependencies. [Masters Thesis]. Delft University of Technology; 2019. Available from: http://resolver.tudelft.nl/uuid:2e793ece-4572-4bb6-83e3-541be467cb4f

University of Minnesota

15. Cai, Feng. Advanced learning approaches based on SVM+ methodology.

Degree: PhD, Electrical Engineering, 2011, University of Minnesota

URL: http://purl.umn.edu/112973

► Exploiting additional information to improve traditional inductive learning is an active research area in machine learning. In many supervised learning applications, training data contains additional… (more)

▼ Exploiting additional information to improve traditional inductive learning is an active research area in machine learning. In many supervised learning applications, training data contains additional information not reflected in training pairs . Examples include: (1) time series prediction where future samples can be observed in the training data, (2) handwritten digit recognition where training examples are provided by several persons, and this group information is not utilized during training, (3) medical diagnosis where predictive (diagnostic) model, say for lung cancer, is estimated using a training set of male and female patients. The gender can be considered as additional group information. Incorporating this additional information into learning may improve generalization. Recently, Vapnik proposed a general approach for incorporating additional information into learning, known as Learning Using Privileged Information (LUPI) and learning with structured data (LWSD) which utilizes group information (Vapnik, 2006). A SVM based methodology SVM+ was proposed under LUPI and LWSD setting (Vapnik, 2006). In this thesis, we will first introduce a SVM+ based feature selection system. Then we extend SVM+ to multi-task learning (MTL) setting, where both training and test data can be naturally partitioned into several groups. SVM+ based MTL (SVM+MTL) method for both classification and regression are proposed and analyzed. SVM+MTL estimates multiple models simultaneously, i.e. one model for each group/task. Task inter-dependency is modeled by sharing a common part of the decision function among different groups. Connections and differences between SVM+ and SVM+MTL are discussed. Practical parameter tuning strategies are proposed for SVM+MTL. Empirical comparisons show that SVM+MTL works very well on data sets with group information. Finally, generalized sequential minimal optimization (GSMO) methods are proposed for SVM+MTL training, for both classification and regression settings.

Subjects/Keywords: feature selection; multi-task learning; SMO; SVM+; SVM+MTL; Electrical Engineering

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Cai, F. (2011). Advanced learning approaches based on SVM+ methodology. (Doctoral Dissertation). University of Minnesota. Retrieved from http://purl.umn.edu/112973

Chicago Manual of Style (16^th Edition):

Cai, Feng. “Advanced learning approaches based on SVM+ methodology.” 2011. Doctoral Dissertation, University of Minnesota. Accessed March 09, 2021. http://purl.umn.edu/112973.

MLA Handbook (7^th Edition):

Cai, Feng. “Advanced learning approaches based on SVM+ methodology.” 2011. Web. 09 Mar 2021.

Vancouver:

Cai F. Advanced learning approaches based on SVM+ methodology. [Internet] [Doctoral dissertation]. University of Minnesota; 2011. [cited 2021 Mar 09]. Available from: http://purl.umn.edu/112973.

Council of Science Editors:

Cai F. Advanced learning approaches based on SVM+ methodology. [Doctoral Dissertation]. University of Minnesota; 2011. Available from: http://purl.umn.edu/112973

Virginia Tech

16. Li, Yifu. Data Filtering and Modeling for Smart Manufacturing Network.

Degree: PhD, Industrial and Systems Engineering, 2020, Virginia Tech

URL: http://hdl.handle.net/10919/99713

► The advancement of the Internet-of-Things (IoT) integrates manufacturing processes and equipment into a network. Practitioners analyze and apply the data generated from the network to… (more)

Subjects/Keywords: Data Filtering; Distributed Filtering and Modeling; Multi-task Learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Li, Y. (2020). Data Filtering and Modeling for Smart Manufacturing Network. (Doctoral Dissertation). Virginia Tech. Retrieved from http://hdl.handle.net/10919/99713

Chicago Manual of Style (16^th Edition):

Li, Yifu. “Data Filtering and Modeling for Smart Manufacturing Network.” 2020. Doctoral Dissertation, Virginia Tech. Accessed March 09, 2021. http://hdl.handle.net/10919/99713.

MLA Handbook (7^th Edition):

Li, Yifu. “Data Filtering and Modeling for Smart Manufacturing Network.” 2020. Web. 09 Mar 2021.

Vancouver:

Li Y. Data Filtering and Modeling for Smart Manufacturing Network. [Internet] [Doctoral dissertation]. Virginia Tech; 2020. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/10919/99713.

Council of Science Editors:

Li Y. Data Filtering and Modeling for Smart Manufacturing Network. [Doctoral Dissertation]. Virginia Tech; 2020. Available from: http://hdl.handle.net/10919/99713

Virginia Tech

17. Nallendran, Vignesh Raja. Predicting Performance Run-time Metrics in Fog Manufacturing using Multi-task Learning.

Degree: MS, Industrial and Systems Engineering, 2021, Virginia Tech

URL: http://hdl.handle.net/10919/102501

► Smart manufacturing aims at utilizing Internet of things (IoT), data analytics, cloud computing, etc. to handle varying market demand without compromising the productivity or quality… (more)

Subjects/Keywords: Fog computing; Fog manufacturing; Multi-task learning; Run-time metrics

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Nallendran, V. R. (2021). Predicting Performance Run-time Metrics in Fog Manufacturing using Multi-task Learning. (Masters Thesis). Virginia Tech. Retrieved from http://hdl.handle.net/10919/102501

Chicago Manual of Style (16^th Edition):

Nallendran, Vignesh Raja. “Predicting Performance Run-time Metrics in Fog Manufacturing using Multi-task Learning.” 2021. Masters Thesis, Virginia Tech. Accessed March 09, 2021. http://hdl.handle.net/10919/102501.

MLA Handbook (7^th Edition):

Nallendran, Vignesh Raja. “Predicting Performance Run-time Metrics in Fog Manufacturing using Multi-task Learning.” 2021. Web. 09 Mar 2021.

Vancouver:

Nallendran VR. Predicting Performance Run-time Metrics in Fog Manufacturing using Multi-task Learning. [Internet] [Masters thesis]. Virginia Tech; 2021. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/10919/102501.

Council of Science Editors:

Nallendran VR. Predicting Performance Run-time Metrics in Fog Manufacturing using Multi-task Learning. [Masters Thesis]. Virginia Tech; 2021. Available from: http://hdl.handle.net/10919/102501

University of Sydney

18. Li, Jizhizi. End-to-end Animal Matting .

Degree: 2020, University of Sydney

URL: http://hdl.handle.net/2123/22897

► Image matting is a widely studied low-level vision problem that aims to provide a detailed foreground alpha matte from a single image, benefiting a wide… (more)

Subjects/Keywords: image matting; animal; segmentation; multi-task; deep learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Li, J. (2020). End-to-end Animal Matting . (Thesis). University of Sydney. Retrieved from http://hdl.handle.net/2123/22897

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16^th Edition):

Li, Jizhizi. “End-to-end Animal Matting .” 2020. Thesis, University of Sydney. Accessed March 09, 2021. http://hdl.handle.net/2123/22897.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7^th Edition):

Li, Jizhizi. “End-to-end Animal Matting .” 2020. Web. 09 Mar 2021.

Vancouver:

Li J. End-to-end Animal Matting . [Internet] [Thesis]. University of Sydney; 2020. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/2123/22897.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Li J. End-to-end Animal Matting . [Thesis]. University of Sydney; 2020. Available from: http://hdl.handle.net/2123/22897

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

UCLA

19. Chen, Wan-Ping. Bayesian Method for Support Union Recovery in Multivariate Multi-Response Linear Regression.

Degree: Statistics, 2015, UCLA

URL: http://www.escholarship.org/uc/item/7174k4ps

► Sparse modeling has become a particularly important and quickly developing topic in many applications of statistics, machine learning, and signal processing. The main objective of… (more)

Subjects/Keywords: Statistics; Bayesian; Multi-response Linear Regression; Multi-task Learning; Support Union Recovery

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Chen, W. (2015). Bayesian Method for Support Union Recovery in Multivariate Multi-Response Linear Regression. (Thesis). UCLA. Retrieved from http://www.escholarship.org/uc/item/7174k4ps

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16^th Edition):

Chen, Wan-Ping. “Bayesian Method for Support Union Recovery in Multivariate Multi-Response Linear Regression.” 2015. Thesis, UCLA. Accessed March 09, 2021. http://www.escholarship.org/uc/item/7174k4ps.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7^th Edition):

Chen, Wan-Ping. “Bayesian Method for Support Union Recovery in Multivariate Multi-Response Linear Regression.” 2015. Web. 09 Mar 2021.

Vancouver:

Chen W. Bayesian Method for Support Union Recovery in Multivariate Multi-Response Linear Regression. [Internet] [Thesis]. UCLA; 2015. [cited 2021 Mar 09]. Available from: http://www.escholarship.org/uc/item/7174k4ps.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Chen W. Bayesian Method for Support Union Recovery in Multivariate Multi-Response Linear Regression. [Thesis]. UCLA; 2015. Available from: http://www.escholarship.org/uc/item/7174k4ps

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

University of Melbourne

20. Karunaratne, Pasan Manura. Scalable and accurate forecasting for smart cities.

Degree: 2018, University of Melbourne

URL: http://hdl.handle.net/11343/214669

► Cities are getting bigger, better and smarter. The increased connectivity of people and devices and the availability of cheap sensors has led to a surge… (more)

▼ Cities are getting bigger, better and smarter. The increased connectivity of people and devices and the availability of cheap sensors has led to a surge in public and government interest in smart city initiatives. This public interest, along with the recent increased interest in machine learning techniques has led to growing research focus into the mining and analysis of data in smart city settings. Much of the analysis in smart city settings is based on forecasting on time series data recorded by smart sensors for planning purposes. For example, utility companies can use electricity load forecasting on smart meter data for capacity planning, and prediction of pedestrian counts and passenger flow in public transportation systems can help in planning to reduce traffic congestion. Though forecasting in smart city settings yields such benefits, it also entails unique challenges, such as challenges related to multi-step prediction, challenges related to low quality training data due to sensors encountering vandalism, malfunction or communication failures, and challenges in maintaining predictive throughput in systems involving increasingly larger numbers of smart sensors. Improving accuracy is a primary goal in any forecasting task, which is especially challenging in multi-step prediction scenarios. We address this challenge by providing new methods to incorporate prior knowledge uniquely relevant to smart cities, such as the periodic behaviour of sensor time series data over the Monday-Friday working week. Specifically, we propose novel kernel function compositions which can incorporate such prior knowledge to kernel-based Bayesian forecasting techniques, with the goal of improving prediction accuracy and robustness to spurious data. We develop our kernel compositions for the state of the art Gaussian Process Regression technique. The new kernel compositions we develop enable prior knowledge relating to multiple periodic effects of the working week (e.g. daily, weekly, holiday effects) and their interactions to be incorporated in the same model. We also provide methods to mitigate the effects of convergence to local optima in the optimisation process over the hyperparameters used in the Gaussian Process models. We address the challenges relating to missing training data in smart city settings by making use of data of other related sensors (which may have more complete data) to mitigate the impact the low quality data has on prediction accuracy. To this end, we develop multi-task learning methods (which are able to learn joint representations from multiple sensors) to improve Gaussian Process Regression prediction accuracy with missing training data values. We also provide equivalent expressions to our multi-task learning methods as combinations of commonly used kernel functions in Gaussian Processes. This enables the straightforward implementation of these methods in popular machine learning toolkits. We address the scalability challenge of large volumes of sensor data in two steps. One, we focus on an…

Subjects/Keywords: smart cities; time series; multi-step forecasting; multi-task learning; gaussian process regression

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Karunaratne, P. M. (2018). Scalable and accurate forecasting for smart cities. (Doctoral Dissertation). University of Melbourne. Retrieved from http://hdl.handle.net/11343/214669

Chicago Manual of Style (16^th Edition):

Karunaratne, Pasan Manura. “Scalable and accurate forecasting for smart cities.” 2018. Doctoral Dissertation, University of Melbourne. Accessed March 09, 2021. http://hdl.handle.net/11343/214669.

MLA Handbook (7^th Edition):

Karunaratne, Pasan Manura. “Scalable and accurate forecasting for smart cities.” 2018. Web. 09 Mar 2021.

Vancouver:

Karunaratne PM. Scalable and accurate forecasting for smart cities. [Internet] [Doctoral dissertation]. University of Melbourne; 2018. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/11343/214669.

Council of Science Editors:

Karunaratne PM. Scalable and accurate forecasting for smart cities. [Doctoral Dissertation]. University of Melbourne; 2018. Available from: http://hdl.handle.net/11343/214669

Carnegie Mellon University

21. Zhang, Yi. Learning with Limited Supervision by Input and Output Coding.

Degree: 2012, Carnegie Mellon University

URL: http://repository.cmu.edu/dissertations/156

► In many real-world applications of supervised learning, only a limited number of labeled examples are available because the cost of obtaining high-quality examples is high.… (more)

▼ In many real-world applications of supervised learning, only a limited number of labeled examples are available because the cost of obtaining high-quality examples is high. Even with a relatively large number of labeled examples, the learning problem may still suffer from limited supervision as the complexity of the prediction function increases. Therefore, learning with limited supervision presents a major challenge to machine learning. With the goal of supervision reduction, this thesis studies the representation, discovery and incorporation of extra input and output information in learning. Information about the input space can be encoded by regularization. We first design a semi-supervised learning method for text classification that encodes the correlation of words inferred from seemingly irrelevant unlabeled text. We then propose a multi-task learning framework with a matrix-normal penalty, which compactly encodes the covariance structure of the joint input space of multiple tasks. To capture structure information that is more general than covariance and correlation, we study a class of regularization penalties on model compressibility. Then we design the projection penalty, which encodes the structure information from a dimension reduction while controlling the risk of information loss. Information about the output space can be exploited by error correcting output codes. Using the composite likelihood view, we propose an improved pairwise coding for multi-label classification, which encodes pairwise label density (as opposed to label comparisons) and decodes using variational methods. We then investigate problemdependent codes, where the encoding is learned from data instead of being predefined. We first propose a multi-label output code using canonical correlation analysis, where predictability of the code is optimized. We then argue that both discriminability and predictability are critical for output coding, and propose a max-margin formulation that promotes both discriminative and predictable codes. We empirically study our methods in a wide spectrum of applications, including document categorization, landmine detection, face recognition, brain signal classification, handwritten digit recognition, house price forecasting, music emotion prediction, medical decision, email analysis, gene function classification, outdoor scene recognition, and so forth. In all these applications, our proposed methods for encoding input and output information lead to significantly improved prediction performance.

Subjects/Keywords: regularization; error-correcting output codes; supervised learning; semi-supervised learning; multi-task learning; multi-label classification; dimensionality reduction; Computer Sciences

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Zhang, Y. (2012). Learning with Limited Supervision by Input and Output Coding. (Thesis). Carnegie Mellon University. Retrieved from http://repository.cmu.edu/dissertations/156

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16^th Edition):

Zhang, Yi. “Learning with Limited Supervision by Input and Output Coding.” 2012. Thesis, Carnegie Mellon University. Accessed March 09, 2021. http://repository.cmu.edu/dissertations/156.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7^th Edition):

Zhang, Yi. “Learning with Limited Supervision by Input and Output Coding.” 2012. Web. 09 Mar 2021.

Vancouver:

Zhang Y. Learning with Limited Supervision by Input and Output Coding. [Internet] [Thesis]. Carnegie Mellon University; 2012. [cited 2021 Mar 09]. Available from: http://repository.cmu.edu/dissertations/156.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Zhang Y. Learning with Limited Supervision by Input and Output Coding. [Thesis]. Carnegie Mellon University; 2012. Available from: http://repository.cmu.edu/dissertations/156

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

University of Kansas

22. Zhang, Jintao. Multi-task and Multi-view Learning for Predicting Adverse Drug Reactions.

Degree: PhD, Information Technology, 2012, University of Kansas

URL: http://hdl.handle.net/1808/18634

► Adverse drug reactions (ADRs) present a major concern for drug safety and are a major obstacle in modern drug development. They account for about one-third… (more)

▼ Adverse drug reactions (ADRs) present a major concern for drug safety and are a major obstacle in modern drug development. They account for about one-third of all late-stage drug failures, and approximately 4% of all new chemical entities are withdrawn from the market due to severe ADRs. Although off-target drug interactions are considered to be the major causes of ADRs, the adverse reaction profile of a drug depends on a wide range of factors such as specific features of drug chemical structures, its ADME/PK properties, interactions with proteins, the metabolic machinery of the cellular environment, and the presence of other diseases and drugs. Hence computational modeling for ADRs prediction is highly complex and challenging. We propose a set of statistical learning models for effective ADRs prediction systematically from multiple perspectives. We first discuss available data sources for protein-chemical interactions and adverse drug reactions, and how the data can be represented for effective modeling. We also employ biological network analysis approaches for deeper understanding of the chemical biological mechanisms underlying various ADRs. In addition, since protein-chemical interactions are an important component for ADRs prediction, identifying these interactions is a crucial step in both modern drug discovery and ADRs prediction. The performance of common supervised learning methods for predicting protein-chemical interactions have been largely limited by insufficient availability of binding data for many proteins. We propose two multi-task learning (MTL) algorithms for jointly predicting active compounds of multiple proteins, and our methods outperform existing states of the art significantly. All these related data, methods, and preliminary results are helpful for understanding the underlying mechanisms of ADRs and further studies. ADRs data are complex and noisy, and in many cases we do not fully understand the molecular mechanisms of ADRs. Due to the noisy and heterogeneous data set available for some ADRs, we propose a sparse multi-view learning (MVL) algorithm for predicting a specific ADR - drug-induced QT prolongation, a major life-threatening adverse drug effect. It is crucial to predict the QT prolongation effect as early as possible in drug development. MVL algorithms work very well when complex data from diverse domains are involved and only limited labeled examples are available. Unlike existing MVL methods that use L2-norm co-regularization to obtain a smooth objective function, we propose an L1-norm co-regularized MVL algorithm for predicting QT prolongation, reformulate the objective function, and obtain its gradient in the analytic form. We optimize the decision functions on all views simultaneously and achieve 3-4 fold higher computational speedup, comparing to previous L2-norm co-regularized MVL methods that alternately optimizes one view with the other views fixed until convergence. L1-norm co-regularization enforces sparsity in the learned mapping functions and hence the results are… Advisors/Committee Members: Huan, Jun (advisor), Vakser, Ilya (cmtemember), Im, Wonpil (cmtemember), Deeds, Eric (cmtemember), Potetz, Brian (cmtemember).

Subjects/Keywords: Bioinformatics; Information technology; adverse drug reaction; boosting; co-regularization; inductive learning; multi-task learning; multi-view learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Zhang, J. (2012). Multi-task and Multi-view Learning for Predicting Adverse Drug Reactions. (Doctoral Dissertation). University of Kansas. Retrieved from http://hdl.handle.net/1808/18634

Chicago Manual of Style (16^th Edition):

Zhang, Jintao. “Multi-task and Multi-view Learning for Predicting Adverse Drug Reactions.” 2012. Doctoral Dissertation, University of Kansas. Accessed March 09, 2021. http://hdl.handle.net/1808/18634.

MLA Handbook (7^th Edition):

Zhang, Jintao. “Multi-task and Multi-view Learning for Predicting Adverse Drug Reactions.” 2012. Web. 09 Mar 2021.

Vancouver:

Zhang J. Multi-task and Multi-view Learning for Predicting Adverse Drug Reactions. [Internet] [Doctoral dissertation]. University of Kansas; 2012. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/1808/18634.

Council of Science Editors:

Zhang J. Multi-task and Multi-view Learning for Predicting Adverse Drug Reactions. [Doctoral Dissertation]. University of Kansas; 2012. Available from: http://hdl.handle.net/1808/18634

Georgia Tech

23. Lu, Jiasen. Visually grounded language understanding and generation.

Degree: PhD, Computer Science, 2020, Georgia Tech

URL: http://hdl.handle.net/1853/62745

► The world around us involves multiple modalities – we see objects, feel texture, hear sounds, smell odors and so on. In order for Artificial Intelligence… (more)

▼ The world around us involves multiple modalities – we see objects, feel texture, hear sounds, smell odors and so on. In order for Artificial Intelligence (AI) to make progress in understanding the world around us, it needs to be able to interpret and reason about multiple modalities. In this thesis, I take steps towards studying how inducing appropriate grounding in deep models improves multi-modal AI capabilities, in the context of vision and language. Specifically, I cover these four tasks: visual question answering, neural image captioning, visual dialog and vision and language pretraining. In visual question answering, we collected a large scale visual question answering dataset and I study various baselines to benchmark these tasks. To jointly reason about image and question, I propose a novel co-attention mechanism that can learn fine-grained grounding to answer the question. In image captioning, I address the model designs for grounded caption generation of a image. A key focus is to extend the model with the ability to know when to look at the image when generating each word. For the words which have explicit visual correspondence, we further proposed a novel approach that reconciles classical slot filling approaches with modern neural captioning approaches. As a result, our model can produce natural language explicitly grounded in entities that object detectors find in the image. In visual dialog, I study both sides of the visual dialog agents – questioner and answerer. For modeling answerer which answers visual questions in dialog, I introduce a novel discriminant perceptual loss that transfers knowledge from a discriminative model a generative model. For modeling questioner, I consider an image guessing game as a test-bed for balancing task performance and language drift. I propose a Dialog without Dialog task, which requires agents to generalize from single round visual question generation with full supervision to a multi-round dialog-based image guessing game without direct language supervision. The proposed visually-grounded dialog models that can adapt to new tasks while exhibiting less linguistic drift. In vision and language pretraining, I study more general models that can learn visual groundings from massive meta-data on the internet. I also explore the multi-task vision and language representation learning. Our results not only show that a single model can perform all 12 vision and language tasks, but also that joint training can lead to improvements in task metric compared to single-task training with the same architecture. Through this work, I demonstrate that inducing appropriate grounding in deep models improves multi-modal AI capabilities. Finally, I briefly discuss the challenges in this domain and the extensions of recent works. Advisors/Committee Members: Parikh, Devi (advisor), Batra, Dhruv (advisor), Corso, Jason J. (advisor), Riedl, Mark Owen (advisor), Hoffman, Judy (advisor).

Subjects/Keywords: Computer vision; Natural language processing; Visual question answering; Multi-task learning; Deep learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Lu, J. (2020). Visually grounded language understanding and generation. (Doctoral Dissertation). Georgia Tech. Retrieved from http://hdl.handle.net/1853/62745

Chicago Manual of Style (16^th Edition):

Lu, Jiasen. “Visually grounded language understanding and generation.” 2020. Doctoral Dissertation, Georgia Tech. Accessed March 09, 2021. http://hdl.handle.net/1853/62745.

MLA Handbook (7^th Edition):

Lu, Jiasen. “Visually grounded language understanding and generation.” 2020. Web. 09 Mar 2021.

Vancouver:

Lu J. Visually grounded language understanding and generation. [Internet] [Doctoral dissertation]. Georgia Tech; 2020. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/1853/62745.

Council of Science Editors:

Lu J. Visually grounded language understanding and generation. [Doctoral Dissertation]. Georgia Tech; 2020. Available from: http://hdl.handle.net/1853/62745

University of Arizona

24. Meyer, Josh. Multi-Task and Transfer Learning in Low-Resource Speech Recognition .

Degree: 2019, University of Arizona

URL: http://hdl.handle.net/10150/634249

► This thesis investigates methods for Acoustic Modeling in Automatic Speech Recog- nition, assuming limited access to training data in the target domain. The Acoustic Models… (more)

Subjects/Keywords: Automatic Speech Recognition; Deep Neural Networks; Low-Resource Languages; Multi-Task Learning; Transfer Learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Meyer, J. (2019). Multi-Task and Transfer Learning in Low-Resource Speech Recognition . (Doctoral Dissertation). University of Arizona. Retrieved from http://hdl.handle.net/10150/634249

Chicago Manual of Style (16^th Edition):

Meyer, Josh. “Multi-Task and Transfer Learning in Low-Resource Speech Recognition .” 2019. Doctoral Dissertation, University of Arizona. Accessed March 09, 2021. http://hdl.handle.net/10150/634249.

MLA Handbook (7^th Edition):

Meyer, Josh. “Multi-Task and Transfer Learning in Low-Resource Speech Recognition .” 2019. Web. 09 Mar 2021.

Vancouver:

Meyer J. Multi-Task and Transfer Learning in Low-Resource Speech Recognition . [Internet] [Doctoral dissertation]. University of Arizona; 2019. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/10150/634249.

Council of Science Editors:

Meyer J. Multi-Task and Transfer Learning in Low-Resource Speech Recognition . [Doctoral Dissertation]. University of Arizona; 2019. Available from: http://hdl.handle.net/10150/634249

University of Pennsylvania

25. Isele, David. Lifelong Reinforcement Learning On Mobile Robots.

Degree: 2018, University of Pennsylvania

URL: https://repository.upenn.edu/edissertations/3290

► Machine learning has shown tremendous growth in the past decades, unlocking new capabilities in a variety of fields including computer vision, natural language processing, and… (more)

Subjects/Keywords: lifelong machine learning; multi-task learning; transfer; Artificial Intelligence and Robotics; Robotics

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Isele, D. (2018). Lifelong Reinforcement Learning On Mobile Robots. (Thesis). University of Pennsylvania. Retrieved from https://repository.upenn.edu/edissertations/3290

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16^th Edition):

Isele, David. “Lifelong Reinforcement Learning On Mobile Robots.” 2018. Thesis, University of Pennsylvania. Accessed March 09, 2021. https://repository.upenn.edu/edissertations/3290.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7^th Edition):

Isele, David. “Lifelong Reinforcement Learning On Mobile Robots.” 2018. Web. 09 Mar 2021.

Vancouver:

Isele D. Lifelong Reinforcement Learning On Mobile Robots. [Internet] [Thesis]. University of Pennsylvania; 2018. [cited 2021 Mar 09]. Available from: https://repository.upenn.edu/edissertations/3290.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Isele D. Lifelong Reinforcement Learning On Mobile Robots. [Thesis]. University of Pennsylvania; 2018. Available from: https://repository.upenn.edu/edissertations/3290

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

University of Adelaide

26. Nekrasov, Vladimir. Semantic Image Segmentation and Other Dense Per-Pixel Tasks: Practical Approaches.

Degree: 2020, University of Adelaide

URL: http://hdl.handle.net/2440/129333

► Computer vision-based and deep learning-driven applications and devices are now a part of our everyday life: from modern smartphones with an ever increasing number of… (more)

▼ Computer vision-based and deep learning-driven applications and devices are now a part of our everyday life: from modern smartphones with an ever increasing number of cameras and other sensors to autonomous vehicles such as driverless cars and self-piloting drones. Even though a large portion of the algorithms behind those systems has been known for ages, the computational power together with the abundance of labelled data were lacking until recently. Now, following the Occam’s razor principle, we should start re-thinking those algorithms and strive towards their further simplification, both to improve our own understanding and expand the realm of their practical applications. With those goals in mind, in this work we will concentrate on a particular type of computer vision tasks that predict a certain quantity of interest for each pixel in the input image – these are so-called dense per-pixel tasks. This choice is not by chance: while there has been a huge amount of works concentrated on per-image tasks such as image classification with levels of performance reaching nearly 100%, dense per-pixel tasks bring a different set of challenges that traditionally require more computational resources and more complicated approaches. Throughout this thesis, our focus will be on reducing these computational requirements and instead presenting simple approaches to build practical vision systems that can be used in a variety of settings – e.g. indoors or outdoors, on low-resolution or high-resolution images, solving a single task or multiple tasks at once, running on modern GPU cards or on embedded devices such as Jetson TX. In the first part of the manuscript we will adapt an existing powerful but slow semantic segmentation network into a faster and competitive one through a manual re-design and analysis of its building blocks. With this approach, we will achieve nearly 3× decrease in the number of parameters and in the runtime of the network with an equally high accuracy. In the second part we then will alter this compact network in order to solve multiple dense per-pixel tasks at once, still in real-time. We will also demonstrate the value of predicting multiple quantities at once, as an example creating a 3D semantic reconstruction of the scene. In the third part, we will move away from the manual design and instead will rely on reinforcement learning to automatically traverse the search space of compact semantic segmentation architectures. While the majority of architecture search methods are computationally extremely expensive even for image classification, we will present a solution that requires only 2 generic GPU cards. Finally, in the last part we will extend our automatic architecture search solution to discover tiny but still competitive networks with less than 300K parameters taking only 1.5MB of a disk space. Advisors/Committee Members: Reid, Ian (advisor), Shen, Chunhua (advisor), School of Computer Science (school).

Subjects/Keywords: Semantic segmentation; deep learning; real-time inference; neural architecture search; multi-task learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Nekrasov, V. (2020). Semantic Image Segmentation and Other Dense Per-Pixel Tasks: Practical Approaches. (Thesis). University of Adelaide. Retrieved from http://hdl.handle.net/2440/129333

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16^th Edition):

Nekrasov, Vladimir. “Semantic Image Segmentation and Other Dense Per-Pixel Tasks: Practical Approaches.” 2020. Thesis, University of Adelaide. Accessed March 09, 2021. http://hdl.handle.net/2440/129333.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7^th Edition):

Nekrasov, Vladimir. “Semantic Image Segmentation and Other Dense Per-Pixel Tasks: Practical Approaches.” 2020. Web. 09 Mar 2021.

Vancouver:

Nekrasov V. Semantic Image Segmentation and Other Dense Per-Pixel Tasks: Practical Approaches. [Internet] [Thesis]. University of Adelaide; 2020. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/2440/129333.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Nekrasov V. Semantic Image Segmentation and Other Dense Per-Pixel Tasks: Practical Approaches. [Thesis]. University of Adelaide; 2020. Available from: http://hdl.handle.net/2440/129333

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Georgia Tech

27. Mehta, Nishant A. On sparse representations and new meta-learning paradigms for representation learning.

Degree: PhD, Computer Science, 2013, Georgia Tech

URL: http://hdl.handle.net/1853/52159

► Given the "right" representation, learning is easy. This thesis studies representation learning and meta-learning, with a special focus on sparse representations. Meta-learning is fundamental to… (more)

Subjects/Keywords: Learning theory; Data-dependent complexity; Luckiness; Dictionary learning; Sparse coding; Lasso; Multi-task learning; Meta-learning; Learning to learn

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Mehta, N. A. (2013). On sparse representations and new meta-learning paradigms for representation learning. (Doctoral Dissertation). Georgia Tech. Retrieved from http://hdl.handle.net/1853/52159

Chicago Manual of Style (16^th Edition):

Mehta, Nishant A. “On sparse representations and new meta-learning paradigms for representation learning.” 2013. Doctoral Dissertation, Georgia Tech. Accessed March 09, 2021. http://hdl.handle.net/1853/52159.

MLA Handbook (7^th Edition):

Mehta, Nishant A. “On sparse representations and new meta-learning paradigms for representation learning.” 2013. Web. 09 Mar 2021.

Vancouver:

Mehta NA. On sparse representations and new meta-learning paradigms for representation learning. [Internet] [Doctoral dissertation]. Georgia Tech; 2013. [cited 2021 Mar 09]. Available from: http://hdl.handle.net/1853/52159.

Council of Science Editors:

Mehta NA. On sparse representations and new meta-learning paradigms for representation learning. [Doctoral Dissertation]. Georgia Tech; 2013. Available from: http://hdl.handle.net/1853/52159

Linköping University

28. Rehman, Obaid Ur. Multi-Task Convolutional Learning for Flame Characterization.

Degree: The Division of Statistics and Machine Learning, 2020, Linköping University

URL: http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-166054

►

This thesis explores multi-task learning for combustion flame characterization i.e to learn different characteristics of the combustion flame. We propose a multi-task convolutional neural… (more)

Subjects/Keywords: Multi task learning; multi task convolutional learning; transfer learning; VGG16; CNN; convolutional neural networks; MTL; MTL CNN; Computer Systems; Datorsystem; Probability Theory and Statistics; Sannolikhetsteori och statistik

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Rehman, O. U. (2020). Multi-Task Convolutional Learning for Flame Characterization. (Thesis). Linköping University. Retrieved from http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-166054

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16^th Edition):

Rehman, Obaid Ur. “Multi-Task Convolutional Learning for Flame Characterization.” 2020. Thesis, Linköping University. Accessed March 09, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-166054.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7^th Edition):

Rehman, Obaid Ur. “Multi-Task Convolutional Learning for Flame Characterization.” 2020. Web. 09 Mar 2021.

Vancouver:

Rehman OU. Multi-Task Convolutional Learning for Flame Characterization. [Internet] [Thesis]. Linköping University; 2020. [cited 2021 Mar 09]. Available from: http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-166054.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Rehman OU. Multi-Task Convolutional Learning for Flame Characterization. [Thesis]. Linköping University; 2020. Available from: http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-166054

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Wayne State University

29. Al-Stouhi, Samir. Learning With An Insufficient Supply Of Data Via Knowledge Transfer And Sharing.

Degree: PhD, Electrical and Computer Engineering, 2013, Wayne State University

URL: https://digitalcommons.wayne.edu/oa_dissertations/829

► As machine learning methods extend to more complex and diverse set of problems, situations arise where the complexity and availability of data presents a… (more)

▼ As machine learning methods extend to more complex and diverse set of problems, situations arise where the complexity and availability of data presents a situation where the information source is not "adequate" to generate a representative hypothesis. Learning from multiple sources of data is a promising research direction as researchers leverage ever more diverse sources of information. Since data is not readily available, knowledge has to be transferred from other sources and new methods (both supervised and un-supervised) have to be developed to selectively share and transfer knowledge. In this dissertation, we present both supervised and un-supervised techniques to tackle a problem where learning algorithms cannot generalize and require an extension to leverage knowledge from different sources of data. Knowledge transfer is a difficult problem as diverse sources of data can overwhelm each individual dataset's distribution and a careful set of transformations has to be applied to increase the relevant knowledge at the risk of biasing a dataset's distribution and inducing negative transfer that can degrade a learner's performance. We give an overview of the issues encountered when the learning dataset does not have a sufficient supply of training examples. We categorize the structure of small datasets and highlight the need for further research. We present an instance-transfer supervised classification algorithm to improve classification performance in a target dataset via knowledge transfer from an auxiliary dataset. The improved classification performance of our algorithm is demonstrated with several real-world experiments. We extend the instance-transfer paradigm to supervised classification with "Absolute Rarity'", where a dataset has an insufficient supply of training examples and a skewed class distribution. We demonstrate one solution with a transfer learning approach and another with an imbalanced learning approach and demonstrate the effectiveness of our algorithms with several real world text and demographics classification problems (among others). We present an unsupervised multi-task clustering algorithm where several small datasets are simultaneously clustered and knowledge is transferred between the datasets to improve clustering performance on each individual dataset and we demonstrate the improved clustering performance with an extensive set of experiments. Advisors/Committee Members: Abhilash Pandya, Chandan K. Reddy.

Subjects/Keywords: Data Mining; Imbalanced Learning; Machine Learning; Multi-Task Learning; Rare Dataset; Transfer Learning; Computer Engineering; Computer Sciences

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Al-Stouhi, S. (2013). Learning With An Insufficient Supply Of Data Via Knowledge Transfer And Sharing. (Doctoral Dissertation). Wayne State University. Retrieved from https://digitalcommons.wayne.edu/oa_dissertations/829

Chicago Manual of Style (16^th Edition):

Al-Stouhi, Samir. “Learning With An Insufficient Supply Of Data Via Knowledge Transfer And Sharing.” 2013. Doctoral Dissertation, Wayne State University. Accessed March 09, 2021. https://digitalcommons.wayne.edu/oa_dissertations/829.

MLA Handbook (7^th Edition):

Al-Stouhi, Samir. “Learning With An Insufficient Supply Of Data Via Knowledge Transfer And Sharing.” 2013. Web. 09 Mar 2021.

Vancouver:

Al-Stouhi S. Learning With An Insufficient Supply Of Data Via Knowledge Transfer And Sharing. [Internet] [Doctoral dissertation]. Wayne State University; 2013. [cited 2021 Mar 09]. Available from: https://digitalcommons.wayne.edu/oa_dissertations/829.

Council of Science Editors:

Al-Stouhi S. Learning With An Insufficient Supply Of Data Via Knowledge Transfer And Sharing. [Doctoral Dissertation]. Wayne State University; 2013. Available from: https://digitalcommons.wayne.edu/oa_dissertations/829

Rochester Institute of Technology

30. Dangi, Shusil. Computational Methods for Segmentation of Multi-Modal Multi-Dimensional Cardiac Images.

Degree: PhD, Chester F. Carlson Center for Imaging Science (COS), 2019, Rochester Institute of Technology

URL: https://scholarworks.rit.edu/theses/10271

► Segmentation of the heart structures helps compute the cardiac contractile function quantified via the systolic and diastolic volumes, ejection fraction, and myocardial mass, representing… (more)

▼ Segmentation of the heart structures helps compute the cardiac contractile function quantified via the systolic and diastolic volumes, ejection fraction, and myocardial mass, representing a reliable diagnostic value. Similarly, quantification of the myocardial mechanics throughout the cardiac cycle, analysis of the activation patterns in the heart via electrocardiography (ECG) signals, serve as good cardiac diagnosis indicators. Furthermore, high quality anatomical models of the heart can be used in planning and guidance of minimally invasive interventions under the assistance of image guidance. The most crucial step for the above mentioned applications is to segment the ventricles and myocardium from the acquired cardiac image data. Although the manual delineation of the heart structures is deemed as the gold-standard approach, it requires significant time and effort, and is highly susceptible to inter- and intra-observer variability. These limitations suggest a need for fast, robust, and accurate semi- or fully-automatic segmentation algorithms. However, the complex motion and anatomy of the heart, indistinct borders due to blood flow, the presence of trabeculations, intensity inhomogeneity, and various other imaging artifacts, makes the segmentation task challenging. In this work, we present and evaluate segmentation algorithms for multi-modal, multi-dimensional cardiac image datasets. Firstly, we segment the left ventricle (LV) blood-pool from a tri-plane 2D+time trans-esophageal (TEE) ultrasound acquisition using local phase based filtering and graph-cut technique, propagate the segmentation throughout the cardiac cycle using non-rigid registration-based motion extraction, and reconstruct the 3D LV geometry. Secondly, we segment the LV blood-pool and myocardium from an open-source 4D cardiac cine Magnetic Resonance Imaging (MRI) dataset by incorporating average atlas based shape constraint into the graph-cut framework and iterative segmentation refinement. The developed fast and robust framework is further extended to perform right ventricle (RV) blood-pool segmentation from a different open-source 4D cardiac cine MRI dataset. Next, we employ convolutional neural network based multi-task learning framework to segment the myocardium and regress its area, simultaneously, and show that segmentation based computation of the myocardial area is significantly better than that regressed directly from the network, while also being more interpretable. Finally, we impose a weak shape constraint via multi-task learning framework in a fully convolutional network and show improved segmentation performance for LV, RV and myocardium across healthy and pathological cases, as well as, in the challenging apical and basal slices in two open-source 4D cardiac cine MRI datasets. We demonstrate the accuracy and robustness of the proposed segmentation methods by comparing the obtained results against the provided gold-standard manual segmentations, as well as with other competing segmentation methods. Advisors/Committee Members: Cristian A. Linte.

Subjects/Keywords: Cardiac cine MRI; Cardiac ultrasound; Convolutional neural network; Image registration; Image segmentation; Multi-task learning

Record Details Similar Records Cite « Share

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6^th Edition):

Dangi, S. (2019). Computational Methods for Segmentation of Multi-Modal Multi-Dimensional Cardiac Images. (Doctoral Dissertation). Rochester Institute of Technology. Retrieved from https://scholarworks.rit.edu/theses/10271

Chicago Manual of Style (16^th Edition):

Dangi, Shusil. “Computational Methods for Segmentation of Multi-Modal Multi-Dimensional Cardiac Images.” 2019. Doctoral Dissertation, Rochester Institute of Technology. Accessed March 09, 2021. https://scholarworks.rit.edu/theses/10271.

MLA Handbook (7^th Edition):

Dangi, Shusil. “Computational Methods for Segmentation of Multi-Modal Multi-Dimensional Cardiac Images.” 2019. Web. 09 Mar 2021.

Vancouver:

Dangi S. Computational Methods for Segmentation of Multi-Modal Multi-Dimensional Cardiac Images. [Internet] [Doctoral dissertation]. Rochester Institute of Technology; 2019. [cited 2021 Mar 09]. Available from: https://scholarworks.rit.edu/theses/10271.

Council of Science Editors:

Dangi S. Computational Methods for Segmentation of Multi-Modal Multi-Dimensional Cardiac Images. [Doctoral Dissertation]. Rochester Institute of Technology; 2019. Available from: https://scholarworks.rit.edu/theses/10271

Open Access Theses and Dissertations