subject:(Graphical model)

University of Alberta

1. Sen, Abhishek. Hand Tracking by Fusion of Color and a Range Sensor.

Degree: MS, Department of Computing Science, 2012, University of Alberta

URL: https://era.library.ualberta.ca/files/pv63g128r

► In this work we have developed a decentralized algorithm for efficient localization and tracking of hands from a sequence of depth and colour images. We… (more)

Subjects/Keywords: hand tracking; graphical model

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Sen, A. (2012). Hand Tracking by Fusion of Color and a Range Sensor. (Masters Thesis). University of Alberta. Retrieved from https://era.library.ualberta.ca/files/pv63g128r

Chicago Manual of Style (16^th Edition):

Sen, Abhishek. “Hand Tracking by Fusion of Color and a Range Sensor.” 2012. Masters Thesis, University of Alberta. Accessed March 25, 2019. https://era.library.ualberta.ca/files/pv63g128r.

MLA Handbook (7^th Edition):

Sen, Abhishek. “Hand Tracking by Fusion of Color and a Range Sensor.” 2012. Web. 25 Mar 2019.

Vancouver:

Sen A. Hand Tracking by Fusion of Color and a Range Sensor. [Internet] [Masters thesis]. University of Alberta; 2012. [cited 2019 Mar 25]. Available from: https://era.library.ualberta.ca/files/pv63g128r.

Council of Science Editors:

Sen A. Hand Tracking by Fusion of Color and a Range Sensor. [Masters Thesis]. University of Alberta; 2012. Available from: https://era.library.ualberta.ca/files/pv63g128r

University of Rochester

2. LaCombe, Jason R. Non-Informative Priors for Structural Inference in Bayesian Networks.

Degree: PhD, 2011, University of Rochester

URL: http://hdl.handle.net/1802/15981

► A fundamental problem in multivariate statistics is the determination of dependency relationships among random variables. Bayesian networks equate the dependency properties of the considered variables… (more)

▼ A fundamental problem in multivariate statistics is the determination of dependency relationships among random variables. Bayesian networks equate the dependency properties of the considered variables with edges in a graph, allowing us a mathematically tractable framework for inferring complex relationships. Thus the original dependency problem reduces to one of model selection, with graphs as the models in consideration. A reasonable solution is to choose the graph that is most likely to have generated the observed data. However as the number of variables begins to exceed the sample size, the traditional statistical method of Maximum Likelihood estimation tends to choose graphs that are overly complex{i.e. graphs with an excessive number of edges. We illustrate this over-tting problem by approximating the expected number of excess edges for a sequence of graphs of increasing size. From a Bayesian perspective, one possible solution to the above problem of over-tting is to introduce a prior on the model-space to penalize complexity. The application of a Bayesian framework provides us not only with a means of selecting a reasonable graph, but also of expressing our condence in our selection through the use of posterior density sampling procedures. Our emphasis is on constructing a prior{the Isomorphism (Iso) prior{which we prove to be unbiased in estimating many important characteristics of a model, such as the number of edges. Furthermore, for a special class of problems we are able to demonstrate several important relationships between the Iso and existing priors. We perform three simulation studies to exemplify the performance of the Iso prior against alternatives, with the results supporting our previous conclusions. Finally we close with a discussion{an exposition on our current and future work{as well a presentation of a general theorem regarding the existence and construction of non-informative priors for more general model-selection problems.

Subjects/Keywords: Bayesian-Networks; Model-Selection; Prior Graphical Model

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

LaCombe, J. R. (2011). Non-Informative Priors for Structural Inference in Bayesian Networks. (Doctoral Dissertation). University of Rochester. Retrieved from http://hdl.handle.net/1802/15981

Chicago Manual of Style (16^th Edition):

LaCombe, Jason R. “Non-Informative Priors for Structural Inference in Bayesian Networks.” 2011. Doctoral Dissertation, University of Rochester. Accessed March 25, 2019. http://hdl.handle.net/1802/15981.

MLA Handbook (7^th Edition):

LaCombe, Jason R. “Non-Informative Priors for Structural Inference in Bayesian Networks.” 2011. Web. 25 Mar 2019.

Vancouver:

LaCombe JR. Non-Informative Priors for Structural Inference in Bayesian Networks. [Internet] [Doctoral dissertation]. University of Rochester; 2011. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/1802/15981.

Council of Science Editors:

LaCombe JR. Non-Informative Priors for Structural Inference in Bayesian Networks. [Doctoral Dissertation]. University of Rochester; 2011. Available from: http://hdl.handle.net/1802/15981

Boston University

3. Kang, Xinyu. Statistical methods for topology inference, denoising, and bootstrapping in networks.

Degree: PhD, Mathematics & Statistics, 2018, Boston University

URL: http://hdl.handle.net/2144/33117

► Quite often, the data we observe can be effectively represented using graphs. The underlying structure of the resulting graph, however, might contain noise and does… (more)

▼ Quite often, the data we observe can be effectively represented using graphs. The underlying structure of the resulting graph, however, might contain noise and does not always hold constant across scales. With the right tools, we could possibly address these two problems. This thesis focuses on developing the right tools and provides insights in looking at them. Specifically, I study several problems that incorporate network data within the multi-scale framework, aiming at identifying common patterns and differences, of signals over networks across different scales. Additional topics in network denoising and network bootstrapping will also be discussed. The first problem we consider is the connectivity changes in dynamic networks constructed from multiple time series data. Multivariate time series data is often non-stationary. Furthermore, it is not uncommon to expect changes in a system across multiple time scales. Motivated by these observations, we in-corporate the traditional Granger-causal type of modeling within the multi-scale framework and propose a new method to detect the connectivity changes and recover the dynamic network structure. The second problem we consider is how to denoise and approximate signals over a network adjacency matrix. We propose an adaptive unbalanced Haar wavelet based transformation of the network data, and show that it is efficient in approximation and denoising of the graph signals over a network adjacency matrix. We focus on the exact decompositions of the network, the corresponding approximation theory, and denoising signals over graphs, particularly from the perspective of compression of the networks. We also provide a real data application on denoising EEG signals over a DTI network. The third problem we consider is in network denoising and network inference. Network representation is popular in characterizing complex systems. However, errors observed in the original measurements will propagate to network statistics and hence induce uncertainties to the summaries of the networks. We propose a spectral-denoising based resampling method to produce confidence intervals that propagate the inferential errors for a number of Lipschitz continuous net- work statistics. We illustrate the effectiveness of the method through a series of simulation studies. Advisors/Committee Members: Kolaczyk, Eric D. (advisor).

Subjects/Keywords: Statistics; Network; Graphical model; Multiscale modeling

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Kang, X. (2018). Statistical methods for topology inference, denoising, and bootstrapping in networks. (Doctoral Dissertation). Boston University. Retrieved from http://hdl.handle.net/2144/33117

Chicago Manual of Style (16^th Edition):

Kang, Xinyu. “Statistical methods for topology inference, denoising, and bootstrapping in networks.” 2018. Doctoral Dissertation, Boston University. Accessed March 25, 2019. http://hdl.handle.net/2144/33117.

MLA Handbook (7^th Edition):

Kang, Xinyu. “Statistical methods for topology inference, denoising, and bootstrapping in networks.” 2018. Web. 25 Mar 2019.

Vancouver:

Kang X. Statistical methods for topology inference, denoising, and bootstrapping in networks. [Internet] [Doctoral dissertation]. Boston University; 2018. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/2144/33117.

Council of Science Editors:

Kang X. Statistical methods for topology inference, denoising, and bootstrapping in networks. [Doctoral Dissertation]. Boston University; 2018. Available from: http://hdl.handle.net/2144/33117

University of New South Wales

4. Liu, Xianghang. New Algorithms for Graphical Models and Their Applications in Learning.

Degree: Computer Science & Engineering, 2015, University of New South Wales

URL: http://handle.unsw.edu.au/1959.4/55080 ; https://unsworks.unsw.edu.au/fapi/datastream/unsworks:36494/SOURCE02?view=true

► Probabilistic graphical models bring together graph theory and probability theory in a powerful formalism for multivariate statistical modelling. Since many machine learning problems involve the… (more)

Subjects/Keywords: statistical inference; machine learning; graphical model

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Liu, X. (2015). New Algorithms for Graphical Models and Their Applications in Learning. (Doctoral Dissertation). University of New South Wales. Retrieved from http://handle.unsw.edu.au/1959.4/55080 ; https://unsworks.unsw.edu.au/fapi/datastream/unsworks:36494/SOURCE02?view=true

Chicago Manual of Style (16^th Edition):

Liu, Xianghang. “New Algorithms for Graphical Models and Their Applications in Learning.” 2015. Doctoral Dissertation, University of New South Wales. Accessed March 25, 2019. http://handle.unsw.edu.au/1959.4/55080 ; https://unsworks.unsw.edu.au/fapi/datastream/unsworks:36494/SOURCE02?view=true.

MLA Handbook (7^th Edition):

Liu, Xianghang. “New Algorithms for Graphical Models and Their Applications in Learning.” 2015. Web. 25 Mar 2019.

Vancouver:

Liu X. New Algorithms for Graphical Models and Their Applications in Learning. [Internet] [Doctoral dissertation]. University of New South Wales; 2015. [cited 2019 Mar 25]. Available from: http://handle.unsw.edu.au/1959.4/55080 ; https://unsworks.unsw.edu.au/fapi/datastream/unsworks:36494/SOURCE02?view=true.

Council of Science Editors:

Liu X. New Algorithms for Graphical Models and Their Applications in Learning. [Doctoral Dissertation]. University of New South Wales; 2015. Available from: http://handle.unsw.edu.au/1959.4/55080 ; https://unsworks.unsw.edu.au/fapi/datastream/unsworks:36494/SOURCE02?view=true

University of Missouri – Columbia

5. Liang, Ye. Bayesian methods on selected topics.

Degree: 2012, University of Missouri – Columbia

URL: http://hdl.handle.net/10355/15884

► Bayesian methods are widely adopted nowadays in statistical analysis. It is especially useful for the statistical inference of complex models or hierarchical models, for which… (more)

Subjects/Keywords: Bayesian statistics; spatial statistics; epidemiology; graphical model

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Liang, Y. (2012). Bayesian methods on selected topics. (Thesis). University of Missouri – Columbia. Retrieved from http://hdl.handle.net/10355/15884

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):

Liang, Ye. “Bayesian methods on selected topics.” 2012. Thesis, University of Missouri – Columbia. Accessed March 25, 2019. http://hdl.handle.net/10355/15884.

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

MLA Handbook (7^th Edition):

Liang, Ye. “Bayesian methods on selected topics.” 2012. Web. 25 Mar 2019.

Vancouver:

Liang Y. Bayesian methods on selected topics. [Internet] [Thesis]. University of Missouri – Columbia; 2012. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/10355/15884.

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

Council of Science Editors:

Liang Y. Bayesian methods on selected topics. [Thesis]. University of Missouri – Columbia; 2012. Available from: http://hdl.handle.net/10355/15884

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

University of California – Irvine

6. Ping, Wei. Learning and Inference in Latent Variable Graphical Models.

Degree: Computer Science, 2016, University of California – Irvine

URL: http://www.escholarship.org/uc/item/7q90z4b5

► Probabilistic graphical models such as Markov random fields provide a powerful framework and tools for machine learning, especially for structured output learning. Latent variables naturally… (more)

▼ Probabilistic graphical models such as Markov random fields provide a powerful framework and tools for machine learning, especially for structured output learning. Latent variables naturally exist in many applications of these models; they may arise from partially labeled data, or be introduced to enrich model flexibility. However, the presence of latent variables presents challenges for learning and inference.For example, the standard approach of using maximum a posteriori (MAP) prediction is complicated by the fact that, in latent variable models (LVMs), we typically want to first marginalize out the latent variables, leading to an inference task called marginal MAP. Unfortunately, marginal MAP prediction can be NP-hard even on relatively simple models such as trees, and few methods have been developed in the literature. This thesis presents a class of variational bounds for marginal MAP that generalizes the popular dual-decomposition method for MAP inference, and enables an efficient block coordinate descent algorithm to solve the corresponding optimization. Similarly, when learning LVMs for structured prediction, it is critically important to maintain the effect of uncertainty over latent variables by marginalization. We propose the marginal structured SVM, which uses marginal MAP inference to properly handle that uncertainty inside the framework of max-margin learning.We then turn our attention to an important subclass of latent variable models, restricted Boltzmann machines (RBMs). RBMs are two-layer latent variable models that are widely used to capture complex distributions of observed data, including as building block for deep probabilistic models. One practical problem in RBMs is model selection: we need to determine the hidden (latent) layer size before performing learning. We propose an infinite RBM model and apply the Frank-Wolfe algorithm to solve the resulting learning problem. The resulting algorithm can be interpreted as inserting a hidden variable into a RBM model at each iteration, to easily and efficiently perform model selection during learning. We also study the role of approximate inference in RBMs and conditional RBMs. In particular, there is a common assumption that belief propagation methods do not work well on RBM-based models, especially for learning. In contrast, we demonstrate that for conditional models and structured prediction, learning RBM-based models with belief propagation and its variants can provide much better results than the state-of-the-art contrastive divergence methods.

Subjects/Keywords: Computer science; Dual-decomposition; Graphical Model; Latent Variable Model; Structured SVM

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Ping, W. (2016). Learning and Inference in Latent Variable Graphical Models. (Thesis). University of California – Irvine. Retrieved from http://www.escholarship.org/uc/item/7q90z4b5

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):

Ping, Wei. “Learning and Inference in Latent Variable Graphical Models.” 2016. Thesis, University of California – Irvine. Accessed March 25, 2019. http://www.escholarship.org/uc/item/7q90z4b5.

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

MLA Handbook (7^th Edition):

Ping, Wei. “Learning and Inference in Latent Variable Graphical Models.” 2016. Web. 25 Mar 2019.

Vancouver:

Ping W. Learning and Inference in Latent Variable Graphical Models. [Internet] [Thesis]. University of California – Irvine; 2016. [cited 2019 Mar 25]. Available from: http://www.escholarship.org/uc/item/7q90z4b5.

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

Council of Science Editors:

Ping W. Learning and Inference in Latent Variable Graphical Models. [Thesis]. University of California – Irvine; 2016. Available from: http://www.escholarship.org/uc/item/7q90z4b5

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

University of Waterloo

7. Princz, Daniel. The CRANE Framework for Simulation Model Workflows.

Degree: 2016, University of Waterloo

URL: http://hdl.handle.net/10012/10230

► CRANE is presented as a flexible framework for linking simulation models and model support tools to form integrated modelling systems for engineering and scientific applications,… (more)

▼ CRANE is presented as a flexible framework for linking simulation models and model support tools to form integrated modelling systems for engineering and scientific applications, evaluated using the scientific workflow approach. CRANE was written using an object-oriented programming language; the separation of its core processing component from its user interface; support for plugins that can be updated and enhanced independent of the framework; and with intuitive user-friendly features and human-readable configuration files. Its strength is its ability to connect to legacy simulation models, whose code cannot be modified, through structured and/or free-format text files. The framework contains an engine that interprets the requirements of simulation models and modelling support tools, and facilitates the flow of data between these components in a simulation workflow. In addition, a user interface provides a familiar graphic interface through which the engine can be configured and monitored during the evaluation of the simulation workflow. A case study was undertaken to demonstrate the ability of CRANE to wrap around, configure, and evaluate two versions of a hydrologic simulation model. Using the default parameter configuration, both versions of the model failed to capture the hydrologic regime of the basin; the modified version of the model only marginally improved the results by redistributing excess meltwater in a presumably more physically based way. The modified version of the model allowed excess meltwater to contribute to ponded storage and infiltrate into soil. By contrast, the original version of the model increased the evaporation rate to account for the excess meltwater. Given the poor overall performance of the model in this particular modelling scenario, the contribution of the modification could not be definitively commented upon. It was concluded that further assessment would be improved by better parameterization of the model. CRANE was used to configure the input files for the model, as well as to execute a simple simulation workflow. Unfortunately, the relative simplicity of the case study did not highlight the more advanced features of the framework. As this is a preliminary introduction of the framework, additional and different types of case studies are recommended, the results from which would identify areas where the framework can continue to be developed and enhanced.

Subjects/Keywords: model calibration; parameter estimation; graphical user interface; simulation model workflow; model-independent software

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Princz, D. (2016). The CRANE Framework for Simulation Model Workflows. (Thesis). University of Waterloo. Retrieved from http://hdl.handle.net/10012/10230

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):

Princz, Daniel. “The CRANE Framework for Simulation Model Workflows.” 2016. Thesis, University of Waterloo. Accessed March 25, 2019. http://hdl.handle.net/10012/10230.

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

MLA Handbook (7^th Edition):

Princz, Daniel. “The CRANE Framework for Simulation Model Workflows.” 2016. Web. 25 Mar 2019.

Vancouver:

Princz D. The CRANE Framework for Simulation Model Workflows. [Internet] [Thesis]. University of Waterloo; 2016. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/10012/10230.

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

Council of Science Editors:

Princz D. The CRANE Framework for Simulation Model Workflows. [Thesis]. University of Waterloo; 2016. Available from: http://hdl.handle.net/10012/10230

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

University of Alberta

8. Eastman, Thomas. A disease classifier for metabolic profiles based on metabolic pathway knowledge.

Degree: MS, Department of Computing Science, 2010, University of Alberta

URL: https://era.library.ualberta.ca/files/fx719n29m

► This thesis presents Pathway Informed Analysis (PIA), a classification method for predicting disease states (diagnosis) from metabolic profile measurements that incorporates biological knowledge in the… (more)

▼ This thesis presents Pathway Informed Analysis (PIA), a classification method for predicting disease states (diagnosis) from metabolic profile measurements that incorporates biological knowledge in the form of metabolic pathways. A metabolic pathway describes a set of chemical reactions that perform a specific biological function. A significant amount of biological knowledge produced by efforts to identify and understand these pathways is formalized in readily accessible databases such as the Kyoto Encyclopedia of Genes and Genomes. PIA uses metabolic pathways to identify relationships among the metabolite concentrations that are measured by a metabolic profile. Specifically, PIA assumes that the class-conditional metabolite concentrations (diseased vs. healthy, respectively) follow multivariate normal distributions. It further assumes that conditional independence statements about these distributions derived from the pathways relate the concentrations of the metabolites to each other. The two assumptions allow for a natural representation of the class-conditional distributions using a type of probabilistic graphical model called a Gaussian Markov Random Field. PIA efficiently estimates the parameters defining these distributions from example patients to produce a classifier. It classifies an undiagnosed patient by evaluating both models to determine the most probable class given their metabolic profile. We apply PIA to a data set of cancer patients to diagnose those with a muscle wasting disease called cachexia. Standard machine learning algorithms such as Naive Bayes, Tree-augmented Naive Bayes, Support Vector Machines and C4.5 are used to evaluate the performance of PIA. The overall classification accuracy of PIA is better than these algorithms on this data set but the difference is not statistically significant. We also apply PIA to several other classification tasks. Some involve predicting various manipulations of the metabolic processes performed in experiments with worms. Other tasks are to classify pigs according to properties of their dietary intake. The accuracy of PIA at these tasks is not significantly better than the standard algorithms.

Subjects/Keywords: metabolic profile; cachexia; graphical model; machine learning; metabolic pathway

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Eastman, T. (2010). A disease classifier for metabolic profiles based on metabolic pathway knowledge. (Masters Thesis). University of Alberta. Retrieved from https://era.library.ualberta.ca/files/fx719n29m

Chicago Manual of Style (16^th Edition):

Eastman, Thomas. “A disease classifier for metabolic profiles based on metabolic pathway knowledge.” 2010. Masters Thesis, University of Alberta. Accessed March 25, 2019. https://era.library.ualberta.ca/files/fx719n29m.

MLA Handbook (7^th Edition):

Eastman, Thomas. “A disease classifier for metabolic profiles based on metabolic pathway knowledge.” 2010. Web. 25 Mar 2019.

Vancouver:

Eastman T. A disease classifier for metabolic profiles based on metabolic pathway knowledge. [Internet] [Masters thesis]. University of Alberta; 2010. [cited 2019 Mar 25]. Available from: https://era.library.ualberta.ca/files/fx719n29m.

Council of Science Editors:

Eastman T. A disease classifier for metabolic profiles based on metabolic pathway knowledge. [Masters Thesis]. University of Alberta; 2010. Available from: https://era.library.ualberta.ca/files/fx719n29m

University of Alberta

9. Zhu, Yunan. Estimating Sparse Graphical Models: Insights Through Simulation.

Degree: MS, Department of Mathematical and Statistical Sciences, 2015, University of Alberta

URL: https://era.library.ualberta.ca/files/4j03d241c

► Graphical models are frequently used to explore networks among a set of variables. Several methods for estimating sparse graphs have been proposed and their theoretical… (more)

Subjects/Keywords: bootstrap; penalized log-likelihood; graphical model; glasso; estimate evaluation

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Zhu, Y. (2015). Estimating Sparse Graphical Models: Insights Through Simulation. (Masters Thesis). University of Alberta. Retrieved from https://era.library.ualberta.ca/files/4j03d241c

Chicago Manual of Style (16^th Edition):

Zhu, Yunan. “Estimating Sparse Graphical Models: Insights Through Simulation.” 2015. Masters Thesis, University of Alberta. Accessed March 25, 2019. https://era.library.ualberta.ca/files/4j03d241c.

MLA Handbook (7^th Edition):

Zhu, Yunan. “Estimating Sparse Graphical Models: Insights Through Simulation.” 2015. Web. 25 Mar 2019.

Vancouver:

Zhu Y. Estimating Sparse Graphical Models: Insights Through Simulation. [Internet] [Masters thesis]. University of Alberta; 2015. [cited 2019 Mar 25]. Available from: https://era.library.ualberta.ca/files/4j03d241c.

Council of Science Editors:

Zhu Y. Estimating Sparse Graphical Models: Insights Through Simulation. [Masters Thesis]. University of Alberta; 2015. Available from: https://era.library.ualberta.ca/files/4j03d241c

University of Adelaide

10. Wang, Zhenhua. Markov random fields with unknown heterogeneous graphs.

Degree: 2014, University of Adelaide

URL: http://hdl.handle.net/2440/98246

► Markov Random Fields have been widely used in computer vision problems, for example image denoising, segmentation and human action recognition. The structure of the graph… (more)

▼ Markov Random Fields have been widely used in computer vision problems, for example image denoising, segmentation and human action recognition. The structure of the graph can be determined using human heuristics or domain knowledge, or can be learned from data when assuming graphs are homogeneous in topology. However, there are many applications for heterogeneous graphs. This research concentrates on estimating heterogeneous graphs and labels simultaneously from the observation. The joint estimation of graphs and labels is formulated into maximising a joint likelihood inference. Unfortunately, these inference problems are generally NP-complete, and we thus develop novel algorithms for effectively and efficiently finding approximate solutions. We also demonstrate how to learn Markov random field parameters from data with our inference techniques. Our contributions are as follows. First, we show that estimating graphs and performing maximum a posteriori inference can be achieved simultaneously by solving a bilinear programming problem, for which we provide a branch and bound solver. Second, we relax the inference problem into a tighter non-convex quadratic programming problem, and propose a convex-concave procedure algorithm to solve the non-convex quadratic programming. Third, we derive the partial-dual of a mixed integer programming relaxation of the inference problem, which admits a scalable message passing-style algorithm. Lastly, we show how to learn the parameters of Markov random fields with the structured max-margin training and the proposed inference algorithms. We evaluate the proposed algorithms on both synthetic data and real applications including human action/activity recognition and semantic image segmentation. Our inference algorithms usually outperform the state-of-the-art. Within our proposed algorithms, the quadratic programming algorithm often performs better than the bilinear programming and the message passing algorithms, while the message passing algorithm is the most efficient. Advisors/Committee Members: van den Hengel, Anton John (advisor), Shi, Qinfeng (advisor), Dick, Anthony Robert (advisor), School of Computer Science (school).

Subjects/Keywords: structured learning; Probablistic Graphical Model; Markov Random Field; unknown graph

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Wang, Z. (2014). Markov random fields with unknown heterogeneous graphs. (Thesis). University of Adelaide. Retrieved from http://hdl.handle.net/2440/98246

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):

Wang, Zhenhua. “Markov random fields with unknown heterogeneous graphs.” 2014. Thesis, University of Adelaide. Accessed March 25, 2019. http://hdl.handle.net/2440/98246.

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

MLA Handbook (7^th Edition):

Wang, Zhenhua. “Markov random fields with unknown heterogeneous graphs.” 2014. Web. 25 Mar 2019.

Vancouver:

Wang Z. Markov random fields with unknown heterogeneous graphs. [Internet] [Thesis]. University of Adelaide; 2014. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/2440/98246.

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

Council of Science Editors:

Wang Z. Markov random fields with unknown heterogeneous graphs. [Thesis]. University of Adelaide; 2014. Available from: http://hdl.handle.net/2440/98246

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

Michigan Technological University

11. Shi, Lufeng. USING PROBABILISTIC GRAPHICAL MODELS TO DRAW INFERENCES IN SENSOR NETWORKS WITH TRACKING APPLICATIONS.

Degree: PhD, Department of Electrical and Computer Engineering, 2014, Michigan Technological University

URL: http://digitalcommons.mtu.edu/etds/752

► Sensor networks have been an active research area in the past decade due to the variety of their applications. Many research studies have been… (more)

▼ Sensor networks have been an active research area in the past decade due to the variety of their applications. Many research studies have been conducted to solve the problems underlying the middleware services of sensor networks, such as self-deployment, self-localization, and synchronization. With the provided middleware services, sensor networks have grown into a mature technology to be used as a detection and surveillance paradigm for many real-world applications. The individual sensors are small in size. Thus, they can be deployed in areas with limited space to make unobstructed measurements in locations where the traditional centralized systems would have trouble to reach. However, there are a few physical limitations to sensor networks, which can prevent sensors from performing at their maximum potential. Individual sensors have limited power supply, the wireless band can get very cluttered when multiple sensors try to transmit at the same time. Furthermore, the individual sensors have limited communication range, so the network may not have a 1-hop communication topology and routing can be a problem in many cases. Carefully designed algorithms can alleviate the physical limitations of sensor networks, and allow them to be utilized to their full potential. Graphical models are an intuitive choice for designing sensor network algorithms. This thesis focuses on a classic application in sensor networks, detecting and tracking of targets. It develops feasible inference techniques for sensor networks using statistical graphical model inference, binary sensor detection, events isolation and dynamic clustering. The main strategy is to use only binary data for rough global inferences, and then dynamically form small scale clusters around the target for detailed computations. This framework is then extended to network topology manipulation, so that the framework developed can be applied to tracking in different network topology settings. Finally the system was tested in both simulation and real-world environments. The simulations were performed on various network topologies, from regularly distributed networks to randomly distributed networks. The results show that the algorithm performs well in randomly distributed networks, and hence requires minimum deployment effort. The experiments were carried out in both corridor and open space settings. A in-home falling detection system was simulated with real-world settings, it was setup with 30 bumblebee radars and 30 ultrasonic sensors driven by TI EZ430-RF2500 boards scanning a typical 800 sqft apartment. Bumblebee radars are calibrated to detect the falling of human body, and the two-tier tracking algorithm is used on the ultrasonic sensors to track the location of the elderly people. Advisors/Committee Members: Jindong Tan.

Subjects/Keywords: Graphical Model; Statistical Inference; Target Tracking; Wireless Sensor Network; Computer Engineering

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Shi, L. (2014). USING PROBABILISTIC GRAPHICAL MODELS TO DRAW INFERENCES IN SENSOR NETWORKS WITH TRACKING APPLICATIONS. (Doctoral Dissertation). Michigan Technological University. Retrieved from http://digitalcommons.mtu.edu/etds/752

Chicago Manual of Style (16^th Edition):

Shi, Lufeng. “USING PROBABILISTIC GRAPHICAL MODELS TO DRAW INFERENCES IN SENSOR NETWORKS WITH TRACKING APPLICATIONS.” 2014. Doctoral Dissertation, Michigan Technological University. Accessed March 25, 2019. http://digitalcommons.mtu.edu/etds/752.

MLA Handbook (7^th Edition):

Shi, Lufeng. “USING PROBABILISTIC GRAPHICAL MODELS TO DRAW INFERENCES IN SENSOR NETWORKS WITH TRACKING APPLICATIONS.” 2014. Web. 25 Mar 2019.

Vancouver:

Shi L. USING PROBABILISTIC GRAPHICAL MODELS TO DRAW INFERENCES IN SENSOR NETWORKS WITH TRACKING APPLICATIONS. [Internet] [Doctoral dissertation]. Michigan Technological University; 2014. [cited 2019 Mar 25]. Available from: http://digitalcommons.mtu.edu/etds/752.

Council of Science Editors:

Shi L. USING PROBABILISTIC GRAPHICAL MODELS TO DRAW INFERENCES IN SENSOR NETWORKS WITH TRACKING APPLICATIONS. [Doctoral Dissertation]. Michigan Technological University; 2014. Available from: http://digitalcommons.mtu.edu/etds/752

Penn State University

12. Park, Minwoo. EFFICIENT MEAN-SHIFT BELIEF PROPAGATION.

Degree: PhD, Computer Science and Engineering, 2010, Penn State University

URL: https://etda.libraries.psu.edu/catalog/11156

► Probabilistic graphical models (PGMs) are widely used in many areas of computer vision and machine learning. Since classic belief propagation is not suitable for a… (more)

▼ Probabilistic graphical models (PGMs) are widely used in many areas of computer vision and machine learning. Since classic belief propagation is not suitable for a continuous state space, sampling-based belief propagation methods have been developed, e.g. non-parametric belief propagation (NBP). However NBP requires a large number of samples and its resampling process is slow, preventing its wide applicability. We are motivated to develop an efficient belief propagation method. Starting with a new heuristic method, mean-shift belief propagation (MSBP) that works iteratively with local weighted samples to infer max-marginals within a large or continuous state space, we further propose a novel data-driven MSBP (DDMSBP) based on recent work of smoothing-based optimization that is even more robust at finding a significant mode of the marginal density. We prove that the inference time for DDMSBP is bilinear in terms of the number of samples and the number of nodes in a graph for arbitrary unary potential functions as long as the pair-wise compatibility functions remain Gaussian. We demonstrate the effectiveness of our novel MSBP and DDMSBP methods through simulation and on five challenging computer vision applications: 1) multi-target tracking, 2) 2D articulated body tracking, 3) 3D deformable neuro-image registration, 4) detection of near-regular patterns in real-world images, and 5) perceptual grouping of 2D lattices in urban scenes. The main contributions of this thesis are: 1) the development and validation of a provable inference algorithm DDMSBP for PGMs with state of the art performance; and 2) a set of novel, robust and efficient solutions to five challenging computer vision problems. These solutions are achieved by incorporating our effective inference methods with group theory and domain knowledge into a graphical model framework. Our proposed inference tools and novel applications have demonstrated the power of MSBP and DDMSBP in diverse, high dimensional computer vision problems especially in computational symmetry.

Subjects/Keywords: MRF; Inference; Graphical Model; Belief Propagation; Computer Vision

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Park, M. (2010). EFFICIENT MEAN-SHIFT BELIEF PROPAGATION. (Doctoral Dissertation). Penn State University. Retrieved from https://etda.libraries.psu.edu/catalog/11156

Chicago Manual of Style (16^th Edition):

Park, Minwoo. “EFFICIENT MEAN-SHIFT BELIEF PROPAGATION.” 2010. Doctoral Dissertation, Penn State University. Accessed March 25, 2019. https://etda.libraries.psu.edu/catalog/11156.

MLA Handbook (7^th Edition):

Park, Minwoo. “EFFICIENT MEAN-SHIFT BELIEF PROPAGATION.” 2010. Web. 25 Mar 2019.

Vancouver:

Park M. EFFICIENT MEAN-SHIFT BELIEF PROPAGATION. [Internet] [Doctoral dissertation]. Penn State University; 2010. [cited 2019 Mar 25]. Available from: https://etda.libraries.psu.edu/catalog/11156.

Council of Science Editors:

Park M. EFFICIENT MEAN-SHIFT BELIEF PROPAGATION. [Doctoral Dissertation]. Penn State University; 2010. Available from: https://etda.libraries.psu.edu/catalog/11156

University of Georgia

13. Sulek, Thaddeus Robert. An application of graphical models to fMRI data using the lasso penalty.

Degree: MS, Statistics, 2017, University of Georgia

URL: http://hdl.handle.net/10724/37188

► In this thesis, we study the graphical lasso method and apply it to functional magnetic resonance imaging (fMRI) data. The graphical lasso method enables one… (more)

Subjects/Keywords: Functional Magnetic Resonance Imaging Data; Graphical Model; Lasso; Regions of Interest

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Sulek, T. R. (2017). An application of graphical models to fMRI data using the lasso penalty. (Masters Thesis). University of Georgia. Retrieved from http://hdl.handle.net/10724/37188

Chicago Manual of Style (16^th Edition):

Sulek, Thaddeus Robert. “An application of graphical models to fMRI data using the lasso penalty.” 2017. Masters Thesis, University of Georgia. Accessed March 25, 2019. http://hdl.handle.net/10724/37188.

MLA Handbook (7^th Edition):

Sulek, Thaddeus Robert. “An application of graphical models to fMRI data using the lasso penalty.” 2017. Web. 25 Mar 2019.

Vancouver:

Sulek TR. An application of graphical models to fMRI data using the lasso penalty. [Internet] [Masters thesis]. University of Georgia; 2017. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/10724/37188.

Council of Science Editors:

Sulek TR. An application of graphical models to fMRI data using the lasso penalty. [Masters Thesis]. University of Georgia; 2017. Available from: http://hdl.handle.net/10724/37188

University of Georgia

14. Sulek, Thaddeus Robert. An application of graphical models to fMRI data using the lasso penalty.

Degree: MS, Statistics, 2017, University of Georgia

URL: http://hdl.handle.net/10724/37031

► In this thesis, we study the graphical lasso method and apply it to functional magnetic resonance imaging (fMRI) data. The graphical lasso method enables one… (more)

Subjects/Keywords: Functional Magnetic Resonance Imaging Data; Graphical Model; Lasso; Regions of Interest

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Sulek, T. R. (2017). An application of graphical models to fMRI data using the lasso penalty. (Masters Thesis). University of Georgia. Retrieved from http://hdl.handle.net/10724/37031

Chicago Manual of Style (16^th Edition):

Sulek, Thaddeus Robert. “An application of graphical models to fMRI data using the lasso penalty.” 2017. Masters Thesis, University of Georgia. Accessed March 25, 2019. http://hdl.handle.net/10724/37031.

MLA Handbook (7^th Edition):

Sulek, Thaddeus Robert. “An application of graphical models to fMRI data using the lasso penalty.” 2017. Web. 25 Mar 2019.

Vancouver:

Sulek TR. An application of graphical models to fMRI data using the lasso penalty. [Internet] [Masters thesis]. University of Georgia; 2017. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/10724/37031.

Council of Science Editors:

Sulek TR. An application of graphical models to fMRI data using the lasso penalty. [Masters Thesis]. University of Georgia; 2017. Available from: http://hdl.handle.net/10724/37031

Carnegie Mellon University

15. Wytock, Matt. Optimizing Optimization: Scalable Convex Programming with Proximal Operators.

Degree: 2016, Carnegie Mellon University

URL: http://repository.cmu.edu/dissertations/785

► Convex optimization has developed a wide variety of useful tools critical to many applications in machine learning. However, unlike linear and quadratic programming, general convex… (more)

Subjects/Keywords: convex optimization; proximal operator; operator splitting; Newton method; sparsity; graphical model

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Wytock, M. (2016). Optimizing Optimization: Scalable Convex Programming with Proximal Operators. (Thesis). Carnegie Mellon University. Retrieved from http://repository.cmu.edu/dissertations/785

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):

Wytock, Matt. “Optimizing Optimization: Scalable Convex Programming with Proximal Operators.” 2016. Thesis, Carnegie Mellon University. Accessed March 25, 2019. http://repository.cmu.edu/dissertations/785.

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

MLA Handbook (7^th Edition):

Wytock, Matt. “Optimizing Optimization: Scalable Convex Programming with Proximal Operators.” 2016. Web. 25 Mar 2019.

Vancouver:

Wytock M. Optimizing Optimization: Scalable Convex Programming with Proximal Operators. [Internet] [Thesis]. Carnegie Mellon University; 2016. [cited 2019 Mar 25]. Available from: http://repository.cmu.edu/dissertations/785.

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

Council of Science Editors:

Wytock M. Optimizing Optimization: Scalable Convex Programming with Proximal Operators. [Thesis]. Carnegie Mellon University; 2016. Available from: http://repository.cmu.edu/dissertations/785

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

University of California – Berkeley

16. Kao, Wei-Chun. Algorithms for Next-Generation High-Throughput Sequencing Technologies.

Degree: Electrical Engineering & Computer Sciences, 2011, University of California – Berkeley

URL: http://www.escholarship.org/uc/item/86b9c87d

► Recent advances of DNA sequencing technologies are allowingresearchers to generateimmense amounts of data in a fast and cost effective fashion, enablinggenome-wide association study and populationgenetic… (more)

▼ Recent advances of DNA sequencing technologies are allowingresearchers to generateimmense amounts of data in a fast and cost effective fashion, enablinggenome-wide association study and populationgenetic research which could not be done a decade ago.There are quite numerous computational challenges arising from thisadvancement, however. Examples include efficient algorithms for processing rawdata generated by sequencing instruments, algorithms for detecting andcorrecting sequencing errors, algorithms for detecting genomevariations from sequence data, just to name a few. Because differentsequencing technologies can have drastically differentcharacteristics, these algorithms often need to be adapted in orderto produce most accurate results.In this thesis, I will address a few of the aforementioned problems. First, Iwill describe two model-based basecalling algorithms for the Illuminasequencing platforms: BayesCall and naiveBayesCall. The novelty of BayesCall algorithmis that it is fully unsupervised, requiring notraining data with known labels, and therefore it is applicable to data without a reference sequence.It also dramatically improves sequencing accuracies.Built upon BayesCall algorithm, naiveBayesCall dramatically improves computationalefficiency by approximating the original model without sacrificingaccuracy. We will also show that improved basecall can have positiveeffects on the downstream sequence analysis, such as the detection ofsingle nucleotide polymorphism and the assembly of novel genomes.In the third chapter, an algorithm, called ECHO, for correcting short-readsequencing errors will be described. The correction algorithm efficiently computesall overlaps between sequencing reads and corrects errorsby using statistical models. Since it does not rely on referencegenomes, ECHO can also be applied to de novo sequencing.Most other error correction algorithms require users to specifykey parameters, but the optimal values for these parameters are unknown tousers and can be hard to specify. Without key parameters beingoptimized, the effectiveness of error correction algorithm couldsometimes be dramatically reduced.Based on statistical models, ECHO optimizes these parametersaccordingly. We will show that ECHO can significantly reducesequence error rates and also facilitate downstream sequenceanalysis. It is also demonstrated that ECHO can be extended to detectheterozygousity from sequencing data.These algorithms are developed in hopes tomake downstream analysis of sequence data easier and ultimatelyfacilitate genome researches.

Subjects/Keywords: Computer science; Bioinformatics; Graphical Model; Illumina; Sequencing; Signal processing

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Kao, W. (2011). Algorithms for Next-Generation High-Throughput Sequencing Technologies. (Thesis). University of California – Berkeley. Retrieved from http://www.escholarship.org/uc/item/86b9c87d

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):

Kao, Wei-Chun. “Algorithms for Next-Generation High-Throughput Sequencing Technologies.” 2011. Thesis, University of California – Berkeley. Accessed March 25, 2019. http://www.escholarship.org/uc/item/86b9c87d.

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

MLA Handbook (7^th Edition):

Kao, Wei-Chun. “Algorithms for Next-Generation High-Throughput Sequencing Technologies.” 2011. Web. 25 Mar 2019.

Vancouver:

Kao W. Algorithms for Next-Generation High-Throughput Sequencing Technologies. [Internet] [Thesis]. University of California – Berkeley; 2011. [cited 2019 Mar 25]. Available from: http://www.escholarship.org/uc/item/86b9c87d.

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

Council of Science Editors:

Kao W. Algorithms for Next-Generation High-Throughput Sequencing Technologies. [Thesis]. University of California – Berkeley; 2011. Available from: http://www.escholarship.org/uc/item/86b9c87d

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

University of Washington

17. Li, Zehang. Bayesian Methods for Graphical Models with Limited Data.

Degree: PhD, 2018, University of Washington

URL: http://hdl.handle.net/1773/43158

► Scientific studies in many fields involve understanding and characterizing dependence relationships among large numbers of variables. This can be challenging in settings where data is… (more)

Subjects/Keywords: Bayesian methods; Graphical model; Spike-and-slab; Verbal Autopsy; Statistics; Statistics

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Li, Z. (2018). Bayesian Methods for Graphical Models with Limited Data. (Doctoral Dissertation). University of Washington. Retrieved from http://hdl.handle.net/1773/43158

Chicago Manual of Style (16^th Edition):

Li, Zehang. “Bayesian Methods for Graphical Models with Limited Data.” 2018. Doctoral Dissertation, University of Washington. Accessed March 25, 2019. http://hdl.handle.net/1773/43158.

MLA Handbook (7^th Edition):

Li, Zehang. “Bayesian Methods for Graphical Models with Limited Data.” 2018. Web. 25 Mar 2019.

Vancouver:

Li Z. Bayesian Methods for Graphical Models with Limited Data. [Internet] [Doctoral dissertation]. University of Washington; 2018. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/1773/43158.

Council of Science Editors:

Li Z. Bayesian Methods for Graphical Models with Limited Data. [Doctoral Dissertation]. University of Washington; 2018. Available from: http://hdl.handle.net/1773/43158

University of Cambridge

18. Pacini, Clare. Inferring condition specific regulatory networks with small sample sizes : a case study in Bacillus subtilis and infection of Mus musculus by the parasite Toxoplasma gondii.

Degree: PhD, 2017, University of Cambridge

URL: https://www.repository.cam.ac.uk/handle/1810/269711 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.744325

► Modelling interactions between genes and their regulators is fundamental to understanding how, for example a disease progresses, or the impact of inserting a synthetic circuit… (more)

▼ Modelling interactions between genes and their regulators is fundamental to understanding how, for example a disease progresses, or the impact of inserting a synthetic circuit into a cell. We use an existing method to infer regulatory networks under multiple conditions: the Joint Graphical Lasso (JGL), a shrinkage based Gaussian graphical model. We apply this method to two data sets: one, a publicly available set of microarray experiments perturbing the gram-positive bacteria Bacillus subtilis under multiple experimental conditions; the second, a set of RNA-seq samples of Mouse (Mus musculus) embryonic fibroblasts (MEFs) infected with different strains of the parasite Toxoplasma gondii. In both cases we infer a subset of the regulatory networks using relatively small sample sizes. For the Bacillus subtilis analysis we focused on the use of these regulatory networks in synthetic biology and found examples of transcriptional units active only under a subset of conditions, this information can be useful when designing circuits to have condition dependent behaviour. We developed methods for large network decomposition that made use of the condition information and showed a greater specificity of identifying single transcriptional units from the larger network using our method. Through annotating these results with known information we were able to identify novel connections and found supporting evidence for a selection of these from publicly available experimental results. Biological data collection is typically expensive and due to the relatively small sample sizes of our MEF data set we developed a novel empirical Bayes method for reducing the false discovery rate when estimating block diagonal covariance matrices. Using these methods we were able to infer regulatory networks for the host infected with either the ME49 or RH strain of the parasite. This enabled the identification of known and novel regulatory mechanisms. The Toxoplasma gondii parasite has shown to subvert host function using similar mechanisms as cancers and through our analysis we were able to identify genes, networks and ontologies associated with cancer, including connections that have not previously been associated with T. gondii infection. Finally a Shiny application was developed as an online resource giving access to the Bacillus subtilis inferred networks with interactive methods for exploring the networks including expansion of sub networks and large network decomposition.

Subjects/Keywords: Gaussian graphical model; Regulatory network; Small sample sizes

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Pacini, C. (2017). Inferring condition specific regulatory networks with small sample sizes : a case study in Bacillus subtilis and infection of Mus musculus by the parasite Toxoplasma gondii. (Doctoral Dissertation). University of Cambridge. Retrieved from https://www.repository.cam.ac.uk/handle/1810/269711 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.744325

Chicago Manual of Style (16^th Edition):

Pacini, Clare. “Inferring condition specific regulatory networks with small sample sizes : a case study in Bacillus subtilis and infection of Mus musculus by the parasite Toxoplasma gondii.” 2017. Doctoral Dissertation, University of Cambridge. Accessed March 25, 2019. https://www.repository.cam.ac.uk/handle/1810/269711 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.744325.

MLA Handbook (7^th Edition):

Pacini, Clare. “Inferring condition specific regulatory networks with small sample sizes : a case study in Bacillus subtilis and infection of Mus musculus by the parasite Toxoplasma gondii.” 2017. Web. 25 Mar 2019.

Vancouver:

Pacini C. Inferring condition specific regulatory networks with small sample sizes : a case study in Bacillus subtilis and infection of Mus musculus by the parasite Toxoplasma gondii. [Internet] [Doctoral dissertation]. University of Cambridge; 2017. [cited 2019 Mar 25]. Available from: https://www.repository.cam.ac.uk/handle/1810/269711 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.744325.

Council of Science Editors:

Pacini C. Inferring condition specific regulatory networks with small sample sizes : a case study in Bacillus subtilis and infection of Mus musculus by the parasite Toxoplasma gondii. [Doctoral Dissertation]. University of Cambridge; 2017. Available from: https://www.repository.cam.ac.uk/handle/1810/269711 ; https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.744325

Oklahoma State University

19. Ding, Yi. Probabilistic Graphic Models for Sports Video Mining: Hybrid Generative-discriminative Approaches.

Degree: School of Electrical & Computer Engineering, 2010, Oklahoma State University

URL: http://hdl.handle.net/11244/7854

► With the development of multimedia and internet technologies, there is a growing interest in video mining research that is to discover knowledge existing in the… (more)

Subjects/Keywords: discriminative; generative; graphical model; machine learning; sports video; video mining

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Ding, Y. (2010). Probabilistic Graphic Models for Sports Video Mining: Hybrid Generative-discriminative Approaches. (Thesis). Oklahoma State University. Retrieved from http://hdl.handle.net/11244/7854

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):

Ding, Yi. “Probabilistic Graphic Models for Sports Video Mining: Hybrid Generative-discriminative Approaches.” 2010. Thesis, Oklahoma State University. Accessed March 25, 2019. http://hdl.handle.net/11244/7854.

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

MLA Handbook (7^th Edition):

Ding, Yi. “Probabilistic Graphic Models for Sports Video Mining: Hybrid Generative-discriminative Approaches.” 2010. Web. 25 Mar 2019.

Vancouver:

Ding Y. Probabilistic Graphic Models for Sports Video Mining: Hybrid Generative-discriminative Approaches. [Internet] [Thesis]. Oklahoma State University; 2010. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/11244/7854.

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

Council of Science Editors:

Ding Y. Probabilistic Graphic Models for Sports Video Mining: Hybrid Generative-discriminative Approaches. [Thesis]. Oklahoma State University; 2010. Available from: http://hdl.handle.net/11244/7854

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

University of Cambridge

20. Pacini, Clare. Inferring condition specific regulatory networks with small sample sizes: a case study in bacillus subtilis and infection of mus musculus by the parasite Toxoplasma gondii .

Degree: 2017, University of Cambridge

URL: https://www.repository.cam.ac.uk/handle/1810/269711

► Modelling interactions between genes and their regulators is fundamental to understanding how, for example a disease progresses, or the impact of inserting a synthetic circuit… (more)

▼ Modelling interactions between genes and their regulators is fundamental to understanding how, for example a disease progresses, or the impact of inserting a synthetic circuit into a cell. We use an existing method to infer regulatory networks under multiple conditions: the Joint Graphical Lasso (JGL), a shrinkage based Gaussian graphical model. We apply this method to two data sets: one, a publicly available set of microarray experiments perturbing the gram-positive bacteria Bacillus subtilis under multiple experimental conditions; the second, a set of RNA-seq samples of Mouse (Mus musculus) embryonic fibroblasts (MEFs) infected with different strains of the parasite Toxoplasma gondii. In both cases we infer a subset of the regulatory networks using relatively small sample sizes. For the Bacillus subtilis analysis we focused on the use of these regulatory networks in synthetic biology and found examples of transcriptional units active only under a subset of conditions, this information can be useful when designing circuits to have condition dependent behaviour. We developed methods for large network decomposition that made use of the condition information and showed a greater specificity of identifying single transcriptional units from the larger network using our method. Through annotating these results with known information we were able to identify novel connections and found supporting evidence for a selection of these from publicly available experimental results. Biological data collection is typically expensive and due to the relatively small sample sizes of our MEF data set we developed a novel empirical Bayes method for reducing the false discovery rate when estimating block diagonal covariance matrices. Using these methods we were able to infer regulatory networks for the host infected with either the ME49 or RH strain of the parasite. This enabled the identification of known and novel regulatory mechanisms. The Toxoplasma gondii parasite has shown to subvert host function using similar mechanisms as cancers and through our analysis we were able to identify genes, networks and ontologies associated with cancer, including connections that have not previously been associated with T. gondii infection. Finally a Shiny application was developed as an online resource giving access to the Bacillus subtilis inferred networks with interactive methods for exploring the networks including expansion of sub networks and large network decomposition.

Subjects/Keywords: Gaussian graphical model; Regulatory network; Small sample sizes

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Pacini, C. (2017). Inferring condition specific regulatory networks with small sample sizes: a case study in bacillus subtilis and infection of mus musculus by the parasite Toxoplasma gondii . (Thesis). University of Cambridge. Retrieved from https://www.repository.cam.ac.uk/handle/1810/269711

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):

Pacini, Clare. “Inferring condition specific regulatory networks with small sample sizes: a case study in bacillus subtilis and infection of mus musculus by the parasite Toxoplasma gondii .” 2017. Thesis, University of Cambridge. Accessed March 25, 2019. https://www.repository.cam.ac.uk/handle/1810/269711.

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

MLA Handbook (7^th Edition):

Pacini, Clare. “Inferring condition specific regulatory networks with small sample sizes: a case study in bacillus subtilis and infection of mus musculus by the parasite Toxoplasma gondii .” 2017. Web. 25 Mar 2019.

Vancouver:

Pacini C. Inferring condition specific regulatory networks with small sample sizes: a case study in bacillus subtilis and infection of mus musculus by the parasite Toxoplasma gondii . [Internet] [Thesis]. University of Cambridge; 2017. [cited 2019 Mar 25]. Available from: https://www.repository.cam.ac.uk/handle/1810/269711.

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

Council of Science Editors:

Pacini C. Inferring condition specific regulatory networks with small sample sizes: a case study in bacillus subtilis and infection of mus musculus by the parasite Toxoplasma gondii . [Thesis]. University of Cambridge; 2017. Available from: https://www.repository.cam.ac.uk/handle/1810/269711

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

University of Toronto

21. Yeo, Alexia. COMBINATORIALLY CONSTRAINED PORTFOLIO OPTIMIZATION USING MESSAGE PASSING ALGORITHMS.

Degree: 2018, University of Toronto

URL: http://hdl.handle.net/1807/89580

►

Portfolio optimization aims to find the optimal investment strategy for a series of assets that results in a minimization of the portfolio's variance. Real life… (more)

Subjects/Keywords: Mixed Integer Programming; Portfolio Optimization; Probabilistic Graphical Model; 0796

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Yeo, A. (2018). COMBINATORIALLY CONSTRAINED PORTFOLIO OPTIMIZATION USING MESSAGE PASSING ALGORITHMS. (Masters Thesis). University of Toronto. Retrieved from http://hdl.handle.net/1807/89580

Chicago Manual of Style (16^th Edition):

Yeo, Alexia. “COMBINATORIALLY CONSTRAINED PORTFOLIO OPTIMIZATION USING MESSAGE PASSING ALGORITHMS.” 2018. Masters Thesis, University of Toronto. Accessed March 25, 2019. http://hdl.handle.net/1807/89580.

MLA Handbook (7^th Edition):

Yeo, Alexia. “COMBINATORIALLY CONSTRAINED PORTFOLIO OPTIMIZATION USING MESSAGE PASSING ALGORITHMS.” 2018. Web. 25 Mar 2019.

Vancouver:

Yeo A. COMBINATORIALLY CONSTRAINED PORTFOLIO OPTIMIZATION USING MESSAGE PASSING ALGORITHMS. [Internet] [Masters thesis]. University of Toronto; 2018. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/1807/89580.

Council of Science Editors:

Yeo A. COMBINATORIALLY CONSTRAINED PORTFOLIO OPTIMIZATION USING MESSAGE PASSING ALGORITHMS. [Masters Thesis]. University of Toronto; 2018. Available from: http://hdl.handle.net/1807/89580

ETH Zürich

22. Perković, Emilija. Graphical characterizations of adjustment sets.

Degree: 2018, ETH Zürich

URL: http://hdl.handle.net/20.500.11850/281392

Subjects/Keywords: Causal inference; Graphical model

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Perković, E. (2018). Graphical characterizations of adjustment sets. (Doctoral Dissertation). ETH Zürich. Retrieved from http://hdl.handle.net/20.500.11850/281392

Chicago Manual of Style (16^th Edition):

Perković, Emilija. “Graphical characterizations of adjustment sets.” 2018. Doctoral Dissertation, ETH Zürich. Accessed March 25, 2019. http://hdl.handle.net/20.500.11850/281392.

MLA Handbook (7^th Edition):

Perković, Emilija. “Graphical characterizations of adjustment sets.” 2018. Web. 25 Mar 2019.

Vancouver:

Perković E. Graphical characterizations of adjustment sets. [Internet] [Doctoral dissertation]. ETH Zürich; 2018. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/20.500.11850/281392.

Council of Science Editors:

Perković E. Graphical characterizations of adjustment sets. [Doctoral Dissertation]. ETH Zürich; 2018. Available from: http://hdl.handle.net/20.500.11850/281392

University of Minnesota

23. Fu, Qiang. Efficient inference algorithms for some probabilistic graphical models.

Degree: PhD, Computer science, 2014, University of Minnesota

URL: http://hdl.handle.net/11299/162960

► The probabilistic graphical model framework provides an essential tool to reason coherently from limited and noisy observations. The framework has been used in an enormous… (more)

▼ The probabilistic graphical model framework provides an essential tool to reason coherently from limited and noisy observations. The framework has been used in an enormous range of application domains, which include: natural language processing, computer vision, bioinformatic, robot navigation and many more. We propose several inference algorithms for some probabilistic graphical models. For Bayesian network graphical models, we focus on the problem of overlapping clustering, where a data point is allowed to belong to multiple clusters. We present an overlapping clustering algo- rithm based on multiplicative mixture models. We analyze a general setting where each component of the multiplicative mixture is from an exponential family, and present an efficient alternating maximization algorithm to learn the model and infer overlap- ping clusters. We also propose a Bayesian Overlapping Subspace Clustering (BOSC) model which is a hierarchical generative model for matrices with potentially overlapping uniform sub-block structures. The BOSC model can also handle matrices with missing entries. We propose an EM-style algorithm based on approximate inference using Gibbs sampling and parameter estimation using coordinate descent for the BOSC model. We propose an EM-style algorithm based on approximate inference using Gibbs sampling and parameter estimation using coordinate descent for the BOSC model. We also consider Markov random field graphical models and address the problem of maximum a posteriori (MAP) inference. We first show that the drought detection problem from the climate science domain can be formulated as a MAP inference problem and propose an automatic drought detection problem. We then present a parallel MAP inference algorithm called Bethe-ADMM based on two ideas: tree-decomposition of the graph and the alternating direction method of multipliers (ADMM). However, unlike the standard ADMM, we use an inexact ADMM augmented with a Bethe-divergence based proximal function, which makes each subproblem in ADMM easy to solve in parallel using the sum-product algorithm. We rigorously prove global convergence of Bethe-ADMM. The proposed algorithm is extensively evaluated on both synthetic and real datasets to illustrate its effectiveness. Further, the parallel Bethe-ADMM is shown to scale almost linearly with increasing number of cores.

Subjects/Keywords: Bayesian network; Graphical model; MAP inference; Markov random field; Overlapping; Clustering

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Fu, Q. (2014). Efficient inference algorithms for some probabilistic graphical models. (Doctoral Dissertation). University of Minnesota. Retrieved from http://hdl.handle.net/11299/162960

Chicago Manual of Style (16^th Edition):

Fu, Qiang. “Efficient inference algorithms for some probabilistic graphical models.” 2014. Doctoral Dissertation, University of Minnesota. Accessed March 25, 2019. http://hdl.handle.net/11299/162960.

MLA Handbook (7^th Edition):

Fu, Qiang. “Efficient inference algorithms for some probabilistic graphical models.” 2014. Web. 25 Mar 2019.

Vancouver:

Fu Q. Efficient inference algorithms for some probabilistic graphical models. [Internet] [Doctoral dissertation]. University of Minnesota; 2014. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/11299/162960.

Council of Science Editors:

Fu Q. Efficient inference algorithms for some probabilistic graphical models. [Doctoral Dissertation]. University of Minnesota; 2014. Available from: http://hdl.handle.net/11299/162960

Virginia Tech

24. Shan, Liang. Joint Gaussian Graphical Model for multi-class and multi-level data.

Degree: PhD, Statistics, 2016, Virginia Tech

URL: http://hdl.handle.net/10919/81412

► Gaussian graphical model has been a popular tool to investigate conditional dependency between random variables by estimating sparse precision matrices. The estimated precision matrices could… (more)

▼ Gaussian graphical model has been a popular tool to investigate conditional dependency between random variables by estimating sparse precision matrices. The estimated precision matrices could be mapped into networks for visualization. For related but different classes, jointly estimating networks by taking advantage of common structure across classes can help us better estimate conditional dependencies among variables. Furthermore, there may exist multilevel structure among variables; some variables are considered as higher level variables and others are nested in these higher level variables, which are called lower level variables. In this dissertation, we made several contributions to the area of joint estimation of Gaussian graphical models across heterogeneous classes: the first is to propose a joint estimation method for estimating Gaussian graphical models across unbalanced multi-classes, whereas the second considers multilevel variable information during the joint estimation procedure and simultaneously estimates higher level network and lower level network. For the first project, we consider the problem of jointly estimating Gaussian graphical models across unbalanced multi-class. Most existing methods require equal or similar sample size among classes. However, many real applications do not have similar sample sizes. Hence, in this dissertation, we propose the joint adaptive graphical lasso, a weighted L1 penalized approach, for unbalanced multi-class problems. Our joint adaptive graphical lasso approach combines information across classes so that their common characteristics can be shared during the estimation process. We also introduce regularization into the adaptive term so that the unbalancedness of data is taken into account. Simulation studies show that our approach performs better than existing methods in terms of false positive rate, accuracy, Mathews correlation coefficient, and false discovery rate. We demonstrate the advantage of our approach using liver cancer data set. For the second one, we propose a method to jointly estimate the multilevel Gaussian graphical models across multiple classes. Currently, methods are still limited to investigate a single level conditional dependency structure when there exists the multilevel structure among variables. Due to the fact that higher level variables may work together to accomplish certain tasks, simultaneously exploring conditional dependency structures among higher level variables and among lower level variables are of our main interest. Given multilevel data from heterogeneous classes, our method assures that common structures in terms of the multilevel conditional dependency are shared during the estimation procedure, yet unique structures for each class are retained as well. Our proposed approach is achieved by first introducing a higher level variable factor within a class, and then common factors across classes. The performance of our approach is evaluated on several simulated networks. We also demonstrate the advantage of our approach using… Advisors/Committee Members: Kim, Inyoung (committeechair), Terrell, George R. (committee member), Deng, Xinwei (committee member), Guo, Feng (committee member).

Subjects/Keywords: Bias Correction; Gaussian graphical model; Heterogeneous classes; Joint adaptive graphical lasso; Joint estimation; Multilevel network; Precision matrix; Unbalanced multi-class.

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Shan, L. (2016). Joint Gaussian Graphical Model for multi-class and multi-level data. (Doctoral Dissertation). Virginia Tech. Retrieved from http://hdl.handle.net/10919/81412

Chicago Manual of Style (16^th Edition):

Shan, Liang. “Joint Gaussian Graphical Model for multi-class and multi-level data.” 2016. Doctoral Dissertation, Virginia Tech. Accessed March 25, 2019. http://hdl.handle.net/10919/81412.

MLA Handbook (7^th Edition):

Shan, Liang. “Joint Gaussian Graphical Model for multi-class and multi-level data.” 2016. Web. 25 Mar 2019.

Vancouver:

Shan L. Joint Gaussian Graphical Model for multi-class and multi-level data. [Internet] [Doctoral dissertation]. Virginia Tech; 2016. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/10919/81412.

Council of Science Editors:

Shan L. Joint Gaussian Graphical Model for multi-class and multi-level data. [Doctoral Dissertation]. Virginia Tech; 2016. Available from: http://hdl.handle.net/10919/81412

University of Washington

25. Zheng, Jinyang. Understanding Online Two-sided Market: An Empirical Perspective.

Degree: PhD, 2017, University of Washington

URL: http://hdl.handle.net/1773/40451

► This dissertation provides a comprehensive investigation of online two-sided markets from empirical perspective. The four essays included propose research questions from four different angles and… (more)

▼ This dissertation provides a comprehensive investigation of online two-sided markets from empirical perspective. The four essays included propose research questions from four different angles and answer those questions by using four correspondingly matched empirical setups. In essay 1, we study the adoption of online two-sided market by investigating how two-sided sales promotion affects drivers’ willingness to use the TNC app and how the TNC develops its optimal promotion strategies accordingly. We find that the substantial value of early promotion not only encourages current usage but also fosters learning that sustains drivers’ use of the app afterward. The results also show that revealed tips from passengers signal low quality of service and that platform cashback to passengers has a positive effect on drivers by increasing drivers’ chances of being rewarded. In essay 2, we investigate the information asymmetry problem of online two-sided market by measuring the effect of platform endorsements and consumer-generated reputation on demand in the online service marketplace and the effect of a “conform or be cast out” policy that is applied by the platform to ensure that sellers have platform refund insurance on the platform-wide quality and competency. We find that individuals have relatively consistent sensitivity to consumer-generated reputation, but different perceptions of platform endorsements, some of which could have a negative effect on demand. Regarding the “conform or be cast out” policy, we find that, even though casting-out reduces the variety of sellers and, thus, decreases platform-wide demand and consumer welfare, the negative effect is offset by sellers’ having platform refund insurance. In essay 3, we study the social influence of online two-sided market by investigating how a review-in-review (RIR) affects rating behavior. We find measurable evidence that individuals rate, in part, to satisfy their expectation of gaining social capital. The average rater weights social capital gain at approximately 20%. In essay 4, we examined the role of the mega player in the online two-sided market by estimating the spillover effects of WeChat on the other top-50 most frequently used apps in China. We find that the spillover effects of WeChat are limited; in fact, only two other only two other apps, Tencent News and Taobao, receive positive spillover effects from WeChat. Advisors/Committee Members: Tan, Yong (advisor).

Subjects/Keywords: Graphical model; Online two-sided market; Structural model; Information technology; Business administration; To Be Assigned

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Zheng, J. (2017). Understanding Online Two-sided Market: An Empirical Perspective. (Doctoral Dissertation). University of Washington. Retrieved from http://hdl.handle.net/1773/40451

Chicago Manual of Style (16^th Edition):

Zheng, Jinyang. “Understanding Online Two-sided Market: An Empirical Perspective.” 2017. Doctoral Dissertation, University of Washington. Accessed March 25, 2019. http://hdl.handle.net/1773/40451.

MLA Handbook (7^th Edition):

Zheng, Jinyang. “Understanding Online Two-sided Market: An Empirical Perspective.” 2017. Web. 25 Mar 2019.

Vancouver:

Zheng J. Understanding Online Two-sided Market: An Empirical Perspective. [Internet] [Doctoral dissertation]. University of Washington; 2017. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/1773/40451.

Council of Science Editors:

Zheng J. Understanding Online Two-sided Market: An Empirical Perspective. [Doctoral Dissertation]. University of Washington; 2017. Available from: http://hdl.handle.net/1773/40451

University of Washington

26. McDavid, Andrew. Statistical Hurdle Models for Single Cell Gene Expression: Differential Expression and Graphical Modeling.

Degree: PhD, 2016, University of Washington

URL: http://hdl.handle.net/1773/36846

► This dissertation describes a set of statistical methods developed for analysis of single cell gene expression. A characteristic of single cell expression is bimodal expression,… (more)

▼ This dissertation describes a set of statistical methods developed for analysis of single cell gene expression. A characteristic of single cell expression is bimodal expression, in which two clusters of expression are present. In any given transcript, the null cluster corresponds to cells without detectable expression (hence a non-zero measurement reflects measurement error) while the signal cluster contains cells with a positive, detectable level of expression. Statistical models that accommodate this characteristic are considered. • In Chapter 1, motivation and history of single cell gene expression is considered. Scientific and statistical questions addressable through single cell expression are discussed, and some statistical frameworks for bulk and single cell expression are described. • In Chapter 2, I consider data generated from replicates of single cells and 100 cell aggregates that were assayed through single cell reverse-transcriptase qPCR (rt-qPCR). In rt-qPCR the null cluster manifests as bona-fide zeros, so expression is characterized by zero-inflation of otherwise continuous values. The average expression from single cells and 100-cell replicates is compared to develop quality control metrics that optimize the single-cell, 100-cell concordance. A Hurdle model is proposed, which accounts for the fact that genes at the single-cell level can be on (and a continuous expression measure is recorded) or dichotomously off (and the recorded expression is zero). Based on this model, I derive a combined likelihood-ratio test for differential expression that incorporates both the discrete and continuous components. This chapter was originally published in McDavid et al. [2013]. • In Chapter 3, I consider application of the hurdle model to single cell RNA sequencing (scRNAseq). In these technologies, the binary zero-inflation described found in rt- qPCR-based assays manifests itself as continuous, bimodal expression, motivating a clustering and thresholding procedure to assign expression to a cluster. The Hurdle model, extended and cast as a vector generalized linear model (vGLM), is provided as an R package named MAST. The cellular detection rate (CDR) is defined as the number of expressed genes found in a cell. It is identified as an important latent factor in single cell experiments, and is argued to measure size and efficiency variations among cells. Gene set enrichment analysis using the Hurdle model, and use of residuals defined through such models are discussed. Parts of this chapter were originally published in Finak et al. [2015], McDavid et al. [2014]. • In Chapter 4, the Hurdle model is generalized to model multivariate dependences between cells, permitting the parametrization of graphical models. A neighborhood selection-based method is proposed to leverage group-l1 penalized regression. Networks estimated on single-cell and multi-cell experiments are contrasted and found to be very distinct. In order to synthesize graphs estimated on transcriptome-scale data, a test for enrichment of connections… Advisors/Committee Members: Gottardo, Raphael (advisor), Drton, Mathias (advisor).

Subjects/Keywords: gene expression; graphical model; hurdle model; RNA sequencing; single cell; zero-inflated; Biostatistics; statistics

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

McDavid, A. (2016). Statistical Hurdle Models for Single Cell Gene Expression: Differential Expression and Graphical Modeling. (Doctoral Dissertation). University of Washington. Retrieved from http://hdl.handle.net/1773/36846

Chicago Manual of Style (16^th Edition):

McDavid, Andrew. “Statistical Hurdle Models for Single Cell Gene Expression: Differential Expression and Graphical Modeling.” 2016. Doctoral Dissertation, University of Washington. Accessed March 25, 2019. http://hdl.handle.net/1773/36846.

MLA Handbook (7^th Edition):

McDavid, Andrew. “Statistical Hurdle Models for Single Cell Gene Expression: Differential Expression and Graphical Modeling.” 2016. Web. 25 Mar 2019.

Vancouver:

McDavid A. Statistical Hurdle Models for Single Cell Gene Expression: Differential Expression and Graphical Modeling. [Internet] [Doctoral dissertation]. University of Washington; 2016. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/1773/36846.

Council of Science Editors:

McDavid A. Statistical Hurdle Models for Single Cell Gene Expression: Differential Expression and Graphical Modeling. [Doctoral Dissertation]. University of Washington; 2016. Available from: http://hdl.handle.net/1773/36846

University of Minnesota

27. Xue, Lingzhou. Regularized Learning of High-dimensional Sparse Graphical Models.

Degree: PhD, 2012, University of Minnesota

URL: http://hdl.handle.net/11299/165775

► High-dimensional graphical models are important tools for characterizing complex interactions within a large-scale system. In this thesis, our emphasis is to utilize the increasingly popular… (more)

Subjects/Keywords: Composite Likelihood; Gaussian Copula; High-dimensional Graphical Model; Ising Model; Nonparametric Correlation; Regularization

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Xue, L. (2012). Regularized Learning of High-dimensional Sparse Graphical Models. (Doctoral Dissertation). University of Minnesota. Retrieved from http://hdl.handle.net/11299/165775

Chicago Manual of Style (16^th Edition):

Xue, Lingzhou. “Regularized Learning of High-dimensional Sparse Graphical Models.” 2012. Doctoral Dissertation, University of Minnesota. Accessed March 25, 2019. http://hdl.handle.net/11299/165775.

MLA Handbook (7^th Edition):

Xue, Lingzhou. “Regularized Learning of High-dimensional Sparse Graphical Models.” 2012. Web. 25 Mar 2019.

Vancouver:

Xue L. Regularized Learning of High-dimensional Sparse Graphical Models. [Internet] [Doctoral dissertation]. University of Minnesota; 2012. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/11299/165775.

Council of Science Editors:

Xue L. Regularized Learning of High-dimensional Sparse Graphical Models. [Doctoral Dissertation]. University of Minnesota; 2012. Available from: http://hdl.handle.net/11299/165775

Brno University of Technology

28. Hraboš, Martin. Simulátor robotického pracoviště .

Degree: 2013, Brno University of Technology

URL: http://hdl.handle.net/11012/54934

► Tato práce se zabýva návrhem a implementací aplikace pro simulaci robotického ramene. Jedná se o skutečný model robota, který je umístněný ve Fakulte informačních technologií… (more)

Subjects/Keywords: simulace; robotika; model; grafické rozhraní; OpenRAVE; Qt; simulation; robotics; model; graphical interface; OpenRAVE; Qt

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Hraboš, M. (2013). Simulátor robotického pracoviště . (Thesis). Brno University of Technology. Retrieved from http://hdl.handle.net/11012/54934

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):

Hraboš, Martin. “Simulátor robotického pracoviště .” 2013. Thesis, Brno University of Technology. Accessed March 25, 2019. http://hdl.handle.net/11012/54934.

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

MLA Handbook (7^th Edition):

Hraboš, Martin. “Simulátor robotického pracoviště .” 2013. Web. 25 Mar 2019.

Vancouver:

Hraboš M. Simulátor robotického pracoviště . [Internet] [Thesis]. Brno University of Technology; 2013. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/11012/54934.

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

Council of Science Editors:

Hraboš M. Simulátor robotického pracoviště . [Thesis]. Brno University of Technology; 2013. Available from: http://hdl.handle.net/11012/54934

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

Texas A&M University

29. Zhang, Lin. Application of Bayesian Hierarchical Models in Genetic Data Analysis.

Degree: 2012, Texas A&M University

URL: http://hdl.handle.net/1969.1/148056

► Genetic data analysis has been capturing a lot of attentions for understanding the mechanism of the development and progressing of diseases like cancers, and is… (more)

▼ Genetic data analysis has been capturing a lot of attentions for understanding the mechanism of the development and progressing of diseases like cancers, and is crucial in discovering genetic markers and treatment targets in medical research. This dissertation focuses on several important issues in genetic data analysis, graphical network modeling, feature selection, and covariance estimation. First, we develop a gene network modeling method for discrete gene expression data, produced by technologies such as serial analysis of gene expression and RNA sequencing experiment, which generate counts of mRNA transcripts in cell samples. We propose a generalized linear model to fit the discrete gene expression data and assume that the log ratios of the mean expression levels follow a Gaussian distribution. We derive the gene network structures by selecting covariance matrices of the Gaussian distribution with a hyper-inverse Wishart prior. We incorporate prior network models based on Gene Ontology information, which avails existing biological information on the genes of interest. Next, we consider a variable selection problem, where the variables have natural grouping structures, with application to analysis of chromosomal copy number data. The chromosomal copy number data are produced by molecular inversion probes experiments which measure probe-specific copy number changes. We propose a novel Bayesian variable selection method, the hierarchical structured variable se- lection (HSVS) method, which accounts for the natural gene and probe-within-gene architecture to identify important genes and probes associated with clinically relevant outcomes. We propose the HSVS model for grouped variable selection, where simultaneous selection of both groups and within-group variables is of interest. The HSVS model utilizes a discrete mixture prior distribution for group selection and group-specific Bayesian lasso hierarchies for variable selection within groups. We further provide methods for accounting for serial correlations within groups that incorporate Bayesian fused lasso methods for within-group selection. Finally, we propose a Bayesian method of estimating high-dimensional covariance matrices that can be decomposed into a low rank and sparse component. This covariance structure has a wide range of applications including factor analytical model and random effects model. We model the covariance matrices with the decomposition structure by representing the covariance model in the form of a factor analytic model where the number of latent factors is unknown. We introduce binary indicators for estimating the rank of the low rank component combined with a Bayesian graphical lasso method for estimating the sparse component. We further extend our method to a graphical factor analytic model where the graphical model of the residuals is of interest. We achieve sparse estimation of the inverse covariance of the residuals in the graphical factor model by employing a hyper-inverse Wishart prior method for a decomposable graph and a Bayesian… Advisors/Committee Members: Mallick, Bani K (advisor), Baladandayuthapani, Veera (advisor), Carroll, Raymond J (committee member), Adams, Garry (committee member).

Subjects/Keywords: covariance estimation; feature selection; graphical network modeling; genetic data analysis; Bayesian hierarchical model

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Zhang, L. (2012). Application of Bayesian Hierarchical Models in Genetic Data Analysis. (Thesis). Texas A&M University. Retrieved from http://hdl.handle.net/1969.1/148056

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, Lin. “Application of Bayesian Hierarchical Models in Genetic Data Analysis.” 2012. Thesis, Texas A&M University. Accessed March 25, 2019. http://hdl.handle.net/1969.1/148056.

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, Lin. “Application of Bayesian Hierarchical Models in Genetic Data Analysis.” 2012. Web. 25 Mar 2019.

Vancouver:

Zhang L. Application of Bayesian Hierarchical Models in Genetic Data Analysis. [Internet] [Thesis]. Texas A&M University; 2012. [cited 2019 Mar 25]. Available from: http://hdl.handle.net/1969.1/148056.

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

Council of Science Editors:

Zhang L. Application of Bayesian Hierarchical Models in Genetic Data Analysis. [Thesis]. Texas A&M University; 2012. Available from: http://hdl.handle.net/1969.1/148056

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

Universidade de Brasília

30. Rodrigo Carvalho. Níveis de cobertura vegetal e qualidade do solo sob pastagem em áreas de cerrado.

Degree: 2007, Universidade de Brasília

URL: http://bdtd.bce.unb.br/tedesimplificado/tde_busca/arquivo.php?codArquivo=2553

►

O geoprocessamento é uma ferramenta de enorme aplicabilidade na pesquisa agropecuária. Modelos gráficos que permitem a identificação de áreas degradadas com rapidez são úteis para… (more)

▼

O geoprocessamento é uma ferramenta de enorme aplicabilidade na pesquisa agropecuária. Modelos gráficos que permitem a identificação de áreas degradadas com rapidez são úteis para evitar que locais em estágio inicial de degradação atinjam níveis economicamente inviáveis de recuperação e não comprometesse a qualidade da água comprometessem a qualidade da água. Os objetivos deste trabalho foram determinar o Índice de Cobertura Vegetal (ICV) através de um Sistema de Informações Geográficas (SIG), quantificar a porosidade total (PT), Diâmetro Médio Ponderado (DMP), Grau de Floculação (GF), Matéria Orgânica (MO) e Capacidade de Troca Catiônica (CTC) de áreas sob pastagem cultivada, com no máximo 8 anos de uso, no Vão-do-Paranã, Goiás, e propor um modelo de comparação, através do Índice de Qualidade do Solo (IQS), entre as pastagens e o cerrado nativo. Os resultados mostraram que a maioria das áreas analisadas apresenta-se em estágio de degradação física e química, sendo que os atributos químicos mostraram-se mais sensíveis às alterações decorrentes da exploração do solo. Os modelos revelaram que apenas uma pequena parte da qualidade do solo pode ser explicada pela cobertura vegetal, e ao mesmo tempo confirmaram o fácil entendimento proporcionado por tais modelos. Os solos com maiores teores de areia foram os que apresentaram menores valores de (IQS).

The geoprocessing is a tool of enormous applicability in the agricultural research. The attainment of models that allow the identification of areas degraded with rapidity would be useful so that the places under anthropic use that if they find in initial period of training of degradation of the soil did not reach impracticable levels of recovery economically and did not compromise the quality of the water. The objectives of this work had been to determine the vegetal covering index through a geographic information system, to quantify the soil porosity, aggregate mean weight diameter, degree of flocculation, organic matter and cation exchange capacity of areas under cultivated pasture, with in the maximum 8 years of use, in the Vão-do-Paranã, Goiás, and to consider a comparison model, through soil quality index, between the cultivated pastures and the native tropical savanna. The results had shown that the models had disclosed that only one small part of the quality of the soil can be explained by the vegetal covering, and at the same time had confirmed the easy proportionate agreement for such models. The majority of the analyzed areas is presented in period of training of physical and chemical degradation, being that the chemical attributes had revealed more sensible to the decurrently alterations of the exploration of the soil. The sandy soils had been the ones that had presented minors values of soil quality index.

Advisors/Committee Members: Marilusa Pinto Coelho Lacerda, Wenceslau J. Goedert, Eiyti Kato, Edson Eyji Sano.

Subjects/Keywords: soil; CIENCIAS AGRARIAS; geoprocessing; tropical savanna; modelo gráfico; degradação do solo; geoprocessamento; pastures; graphical model

Record Details Similar Records Cite « Share

❌

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

APA (6^th Edition):

Carvalho, R. (2007). Níveis de cobertura vegetal e qualidade do solo sob pastagem em áreas de cerrado. (Thesis). Universidade de Brasília. Retrieved from http://bdtd.bce.unb.br/tedesimplificado/tde_busca/arquivo.php?codArquivo=2553

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):

Carvalho, Rodrigo. “Níveis de cobertura vegetal e qualidade do solo sob pastagem em áreas de cerrado.” 2007. Thesis, Universidade de Brasília. Accessed March 25, 2019. http://bdtd.bce.unb.br/tedesimplificado/tde_busca/arquivo.php?codArquivo=2553.

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

MLA Handbook (7^th Edition):

Carvalho, Rodrigo. “Níveis de cobertura vegetal e qualidade do solo sob pastagem em áreas de cerrado.” 2007. Web. 25 Mar 2019.

Vancouver:

Carvalho R. Níveis de cobertura vegetal e qualidade do solo sob pastagem em áreas de cerrado. [Internet] [Thesis]. Universidade de Brasília; 2007. [cited 2019 Mar 25]. Available from: http://bdtd.bce.unb.br/tedesimplificado/tde_busca/arquivo.php?codArquivo=2553.

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

Council of Science Editors:

Carvalho R. Níveis de cobertura vegetal e qualidade do solo sob pastagem em áreas de cerrado. [Thesis]. Universidade de Brasília; 2007. Available from: http://bdtd.bce.unb.br/tedesimplificado/tde_busca/arquivo.php?codArquivo=2553

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

		in
And / Or
		in
And / Or
		in
And / Or
		in

Open Access Theses and Dissertations