CS Theses & Dissertations 2023

For 2023 graduation dates (in alphabetical order by last name):

M-NeRF: model-based human reconstruction from scratch with mirror-aware neural radiance fields
Ajisafe, Daniel Abidemi
DOI : 10.14288/1.0423218
URI : http://hdl.handle.net/2429/83704
Degree : Master of Science – MSc
Graduation Date : 2023-05
Supervisor : Dr. Helge Rhodin

Extract : informing microservice extraction decisions from the bottom-up
Akamoto, Shizuko
DOI : 10.14288/1.0435520
URI : http://hdl.handle.net/2429/85558
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Reid Holmes

Provenance in relational databases: usability and applications
AlOmeir, Omar
DOI : 10.14288/1.0430326
URI : http://hdl.handle.net/2429/84156
Degree : Doctor of Philosophy - PhD
Graduation 
Date : 2023-05
Supervisor : Dr. Rachel Pottinger

Data provenance is any information about the origin of a piece of data and the process that led to its creation. Most database provenance work has focused on creating models and semantics to query and generate this provenance information. While comprehensive, provenance information remains large and overwhelming, which can make it hard for data provenance systems to support data exploration or any meaningful applications. This thesis is focused on facilitating the use of database provenance through visual interfaces, summarization techniques, and curation techniques for real world applications. In the first part, we present visualization techniques for provenance information in relational databases. Our visualizations address every part of provenance information to facilitate user exploration. Through a user experiment, we show that our approach improves the accuracy and efficiency of performing exploration tasks. The next part addresses the challenge of volume of provenance information. Specifically, in the case of aggregation queries. The volume increases with the size of the database and creates a “needle in a haystack problem”. We present novel summarization techniques that build on existing summarization literature. Our techniques work to support exploration for users who are not familiar with the data or its provenance. The final part shows our use of our summarization techniques to address the problem of refining aggregate queries. Aggregate queries pose a chaliii lenge in that they present ambiguous results to inexperienced users. Query refinement can help users realize their query errors and help them fix them. Through user experiment, we present evidence of the usefulness, and usability of our methods. Overall, the goal of this thesis is to facilitate the use of provenance information in relational databases. Through the use of novel techniques and user-centric evaluation, we present novel solutions and user interaction methods to enable new applications in this domain. 

Anomaly Detection in Multiplex Networks: From Human Brain Activity to Financial Networks
Behrouz, Ali
DOI : 10.14288/1.0435270
URI : http://hdl.handle.net/2429/85514
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Margo Seltzer

Using transformers to predict customer satisfaction for live chat dialogues: guiding applied natural language processing research in contact centres through design thinking
Boutet, Patrick
DOI : 10.14288/1.0422968
URI : http://hdl.handle.net/2429/83573
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Giuseppe Carenini

Kernel methods for invariant representation learning: enforcing fairness and conditional independence
Deka, Namrata
DOI : 10.14288/1.0431526
URI : http://hdl.handle.net/2429/84530
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Danica Sutherland

On the design of a gradual dependently typed language for programming
Eremondi, Joseph S.
DOI : 10.14288/1.0428823
URI : http://hdl.handle.net/2429/84109
Degree : Doctor of Philosophy - PhD
Graduation Date : 2023-05
Supervisor : Dr. Ron Garcia

Dependently typed programming languages provide a way to write programs, specifications, and correctness proofs using a single language. If a dependent type checker accepts a program, the programmer can be assured that it behaves according to the specification given in its types. However, dependently typed programming languages can be hard to use. Gradual types provide a way to mix dynamically and statically typed code in a single language. Under this paradigm, programs may have imprecise types, causing certain type checks to be deferred to run time. We build the theoretical foundations for combining gradual and dependent types in a programming language, with the aim of making dependent types easier to use. The differences between these two paradigms lead to inherent tensions when choosing the properties such a language should satisfy. Gradual typing's effectful nature conflicts with the compile-time reductions of dependent type checking. Gradual run-time type comparisons clash with dependent types containing terms that bind variables. This dissertation identifies such tensions and proposes a design that finds balance between the conflicting goals. Our contribution has three parts: First, we present a foundational calculus for gradual dependent types, with functions, function types and universes. To ensure that type checking terminates, we reduce compile-time terms with approximate normalization, producing imprecise results when the available type information cannot guarantee termination. We use hereditary substitution to show that approximate normalization always terminates. Second, we present a notion of propositional equality for gradual dependent types. We devise a method of tracking run-time consistency information between imprecise equated terms, and introduce a composition operator in the language itself. Third, we show that the first and second contributions can be combined, giving a language with approximate normalization that supports inductive types and propositional equality with dynamic consistency tracking. Since hereditary substitution does not scale to inductive types, we use a syntactic model to establish termination. The same technique is used to model non-terminating run-time semantics using guarded type theory, paving the road for mechanizing the metatheory of gradual dependent types.

Representation learning with explicit and implicit graph structures
Fatemi, Bahare
DOI : 10.14288/1.0431397
URI : http://hdl.handle.net/2429/84434
Degree : Doctor of Philosophy - PhD
Graduation Date : 2023-05
Supervisor : Dr. David Poole

The world around us is composed of objects each having relations with other objects. The objects and relations form a (hyper)graph with objects as nodes and relations between objects as (hyper)edges. When learning, the underlying structure representing the relations between the nodes is either given explicitly in the training set or is implicit and needs to be inferred. This dissertation studies graph representation learning with both explicit and implicit structures. For explicit structure, we first tackle the challenge of enforcing taxonomic information while embedding entities and relations. We prove that some fully expressive models cannot respect subclass and subproperty information. With minimal modifications to an existing knowledge graph completion method, we enable the injection of taxonomic information. A second challenge is in representing explicit structures in relational hypergraphs that contain relations defined on an arbitrary number of entities. While techniques, such as reification, exist that convert non-binary relations into binary ones, we show that current embedding-based methods do not work well out of the box for knowledge graphs obtained through these techniques. We introduce embedding-based methods that work directly with relations of arbitrary arity. We also develop public datasets, benchmarks, and baselines and show experimentally that the proposed models are more effective than the baselines. We further bridge the gap between relational algebra and knowledge hypergraphs by proposing an embedding-based model that can represent relational algebra operations. Having introduced novel architectures for explicitly graph-structured data, we further investigate how models with relational inductive biases can be developed and applied to problems with implicit structures. Graph representation learning models work well when the structure is explicit. However, this structure may not always be available in real-world applications. We propose the Simultaneous Learning of Adjacency and graph neural network Parameters with Self-supervision, or SLAPS, a method that provides more supervision for inferring a graph structure through self-supervision. An experimental study demonstrates that SLAPS scales to large graphs with hundreds of thousands of nodes and outperforms several baselines on established benchmarks.

A study of the edge of stability in deep learning
Fox, Curtis
DOI : 10.14288/1.0435607
URI : http://hdl.handle.net/2429/85651
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Mark Schmidt 

Imitating optimized trajectories for dynamic quadruped behaviors
Fuchioka, Yuni
DOI : 10.14288/1.0431193
URI : http://hdl.handle.net/2429/84355
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Michiel van de Panne

Neural Multimodal Topic Modeling: A Comprehensive Evaluation
González Pizarro, Felipe
DOI : 10.14288/1.0435191
URI : http://hdl.handle.net/2429/85455
Degree : Master of Science – MSc
Graduation Date : 2023-11
Supervisor : Dr. Giuseppe Carenini

Classifying long-term traits from action and eye-tracking data for personalized XAI in an intelligent tutoring system
Graham, Liam
DOI : 10.14288/1.0423553
URI : http://hdl.handle.net/2429/83744
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Cristina Conati

Simpler specifications for resource-manipulating programs
Grannan, Zachary
DOI : 10.14288/1.0431600
URI : http://hdl.handle.net/2429/84556
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Alexander Summers

Evaluating the quality of student-written software tests with curated mutation analysis
Hall, Braxton
DOI : 10.14288/1.0421806
URI : http://hdl.handle.net/2429/83142
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Reid Holmes & Dr. Elisa Baniassad

Classification of Alzheimer's using deep-learning methods on webcam-based gaze data
Harisinghani, Anuj
DOI : 10.14288/1.0427394
URI : http://hdl.handle.net/2429/83892
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Cristina Conati

Map conflation via knowledge graph representations of digital maps
Hashemi Fesharaki, Seyedeh Farnoosh
DOI : 10.14288/1.0434669
URI :http://hdl.handle.net/2429/85388
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Laks Lakshmanan

GlueFL: Reconciling Client Sampling and Model Masking for Bandwidth Efficient Federated Learning
He, Shiqi
DOI : 10.14288/1.0423118
URI : http://hdl.handle.net/2429/83666
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Ivan Beschastnikh

Self-supervision through Random Segments with Autoregressive Coding (RandSAC)
Hua, Tianyu
DOI : 10.14288/1.0431370
URI : http://hdl.handle.net/2429/84411
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Leonid Sigal

Layered controllable video generation
Huang, Jiahui
DOI : 10.14288/1.0422616
URI : http://hdl.handle.net/2429/83427
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Leonid Sigal

Privacy and conflicting identities in the context of Punjabi Canadians
Iqbal, Faqia
DOI : 10.14288/1.0422997
URI : http://hdl.handle.net/2429/83600
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Ivan Beschastnikh

A Kernel-Based Approach to Differentially Private Image Generation
Jalali Asadabadi, Milad 
DOI : 10.14288/1.0437191
URI : http://hdl.handle.net/2429/86181
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Danica Sutherland

Design opportunities for personalized proactive notification management
Janzen, Izabelle Foster
DOI : 10.14288/1.0422492
URI : http://hdl.handle.net/2429/83396
Degree : Doctor of Philosophy - PhD
Graduation Date : 2023-05
Supervisor : Dr. Joanna McGrenere

Mobile devices deliver ‘notifications’ that interrupt the user for events like receiving text messages and emails. However, frequent interruptions reduce productivity and can increase stress. Notification management tools can reduce this distraction but substantial barriers, such as time and effort, prevent most users from managing their notifications. In this dissertation, I present three projects that aim to understand how to better support proactive notification management and I systematically design a novel notification management tool to better address user needs. We used a qualitative, user-centered approach consisting of interviews (N=71) and surveys (N=140). First, we explored how the notification management needs of older adults compared to the general population because other forms of distraction disproportionately impact older adults but their notification management needs had not been studied. Through semi-structured interviews (N=20) with both younger and older adults, we identified that many users desire a sense of control and agency over their attention that was difficult to achieve with current notification management technology. Second, we characterized how users could proactively control notifications through four design axes. We created five notification management design concepts that included design elements from under-explored areas of our design axes. Through semi-structured interviews (N=30), we probed user reactions to our design concepts to identify what design elements help address user concerns, such as a lack of control in notification management. The results of this study suggested there was strong interest to include reflective design elements within notification management. Lastly, we leveraged these results and a supplementary survey of users’ notification preferences (N=140) to design a novel notification management tool. The Reflective Spring Cleaning design supported infrequent notification management with visualizations of notification usage data and suggestions for personalization. We evaluated this design through a longitudinal study organized around semi-structured interviews (N=21). Our results demonstrated how the concept encouraged and supported users to proactively manage notifications through critical reflection on the impact of notifications. Our work outlines key design directions for how notification management tools can support proactive notification management and provide users a sense of control over their attention.

[no title]
Jhaj, Arshvir
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Danica Sutherland

Modularizing deep learning for geometry-aware registration and reconstruction
Jiang, Wei
DOI : 10.14288/1.0427395
URI : http://hdl.handle.net/2429/83895
Degree : Doctor of Philosophy – PhD
Graduation Date : 2023-05
Supervisor : Dr. Kwang Moo Yi

In this work, we explore the modularization of deep learning for geometry-aware registration and reconstruction, with a particular focus on cameras registration and human reconstruction from videos. The traditional methods for these tasks have been challenged by deep learning approaches, but end-to-end learning can be limited in terms of generalization, transparency, and controllability. Modularization breaks the task into smaller subtasks and allows each to be addressed individually using traditional methods or deep learning techniques. Through modularization, we are able to embed knowledge from the real world, enabling better generalization, simpler and more effective learning, explainable and transparent models, and geometry-awareness. Specifically, this work consists of four major chapters, each presenting a modularized approach to solve a specific geometric problem. Firstly, a novel linearized multi-sampling method is proposed to enable better image alignment and learning. Secondly, the homography warping is modularized out of the pipeline allowing optimization through the learned error for accurate sports field registration. Thirdly, by modularizing the robust estimation and 3D map from the pose estimation pipeline, the neural network can focus on learning accurate image correspondences. Finally, the modularization of human scene positioning and mesh skinning allows for the reconstruction of animatable human avatar from video. Overall, our work demonstrates the power of modularization, and we hope it will inspire future research on modularization and its potential applications to other areas.

[no title]
Jiao, Yibo
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Dinesh Pai

Designing integrated development environments for all ages through tinkering
Kerr, Katharine
DOI : 10.14288/1.0435606
URI : http://hdl.handle.net/2429/85643
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Reid Holmes

Constrained dynamics with frictional contact on smooth surfaces
Larionov, Egor
DOI : 10.14288/1.0422992
URI : http://hdl.handle.net/2429/83599
Degree : Doctor of Philosophy - PhD
Graduation Date : 2023-05
Supervisor : Dr. Dinesh Pai

Friction and contact pose a great challenge to efficient and accurate simulation of deformable objects for computer graphics and engineering applications. In contrast to many engineering applications, simulation software for graphics often permits larger approximation errors in favour of better predictability, controllability and efficiency. This dissertation explores modern methods for frictional contact resolution in computer graphics. In particular, the focus is on offline simulation of smooth elastic objects subject to contact with other elastic solids and cloth. We explore traditional non-smooth friction formulations as well as smoothed frictional contact, which lends itself well to differentiable simulation and analysis. We then explore a particular application of differentiable simulation to motivate the direction of research. In graphics, even smooth objects are typically approximated using piecewise linear polyhedra, which exhibit sliding artifacts that can be interpreted as artificial friction making simulations less predictable. We develop a technique for improving fidelity of sliding contact between smooth objects. Frictional contacts are traditionally resolved using non-smooth models, which are complex to analyse and difficult to compute to a desirable error estimate. We propose a unified description of the equations of motion subject to frictional contacts using a smooth model that converges to an accurate friction response. We further analyse the implications of this formulation and compare our results to state-of-the-art methods. The smooth model uniquely resolves frictional contacts, while also being fully differentiable. This allows inverse problems using our formulation to be solved by gradient-based methods. We begin our exploration of differentiable simulation applications with a parameter estimation task. Elastic parameters are estimated for a three distinct cloth materials using a novel capture, registration and estimation pipeline. Static equilibrium cloth configurations are efficiently estimated using a popular compliant constraint dynamics. In this work we address a common issue of bifurcation in cloth, which causes final configuration mismatches during estimation. Finally, we postulate an extension to compliant constraint dynamics using our friction model, to show how our previous work can be used in parameter estimation tasks involving contact and friction.

Computationally efficient geometric methods for optimization and inference in machine learning
Lin, Wu
DOI : 10.14288/1.0434160
URI : http://hdl.handle.net/2429/85203
Degree : Doctor of Philosophy – PhD
Graduation Date : 2023-11
Supervisor : Dr. Mark Schmidt & Dr. Mohammad Emtiyaz Khan

Optimization and inference are essential ingredients of machine learning (ML). Many optimization and inference problems can be formulated from a probabilistic perspective to exploit the Fisher-Rao geometric structure of a probability family. By leveraging the structure, optimization and inference methods can improve their performance. A classic approach to exploiting the Fisher-Rao structure is natural-gradient descent (NGD). However, implementing NGD remains computationally intensive. When a parameter space is constrained, it can also be non-trivial to perform NGD while taking the constraint into consideration. It is even more challenging to perform NGD in structured parameter spaces. This work pushes the boundary of the theory and practice of NGD. We first establish a connection among NGD, mirror descent (MD), and Riemannian gradient descent (RGD). Thanks to this connection, we consider various approaches to better exploit underlying geometric and algebraic structures of NGD resulting in computationally efficient and practically useful methods for inference and optimization problems in ML. In the first part of this work, we reduce the computational cost of NGD for exponential family approximations in variational inference. To do so, we exploit a convex duality in MD by viewing NGD as a MD update. Secondly, we extend the scope of NGD and develop efficient methods by exploiting a similar duality for mixtures of exponential family approximations. Mixtures are more powerful than the exponential family to model complex distributions. We introduce a new Fisher-Rao metric for a mixture since the original Fisher-Rao metric becomes singular in mixture cases. The third part of this work addresses the constraint issue in NGD by proposing a new efficient update scheme inspired by RGD. We focus on performing NGD updates in positive-definite constrained parameter spaces as these spaces appear in exponential family distributions and their mixtures. Finally, we enable a practical usage of NGD methods in structured and constrained parameter spaces by using tools from Riemannian geometry. We propose a structure-preserving NGD update scheme by using matrix Lie groups and moving coordinates. Thanks to our approach, we also develop novel structured second-order methods for unconstrained optimization and adaptive gradient-based methods for deep learning.

Processing Freehand Vector Sketches
Liu, Chenxi
DOI : 10.14288/1.0433727
URI : http://hdl.handle.net/2429/85133
Degree : Doctor of Philosophy - PhD
Graduation Date : 2023-11
Supervisor : Dr. Alla Sheffer

Freehand sketching is a fast and intuitive way for artists to communicate visual ideas, and is often the first step of creating visual content, ranging from industrial design to cartoon production. As drawing tablets and touch displays become increasingly common among professionals, a growing number of sketches are created and stored digitally in vector graphics format. This trend motivates a series of downstream sketch-based applications, performing tasks including drawing colorization, 3D model creation, editing, and posing. Even when stored digitally in vector format, hand-drawn sketches, often containing overdrawn strokes and inaccurate junctions, are different from the clean vector sketches required by these applications, which results in tedious and time-consuming manual cleanup tasks. In this thesis, we analyze the human perceptual cues that influence these two tasks: grouping overdrawn strokes that depict a single intended curve and connecting unintended gaps between strokes. Guided by these cues, we develop three methods for these two tasks. We first introduce StrokeAggregator, a method that automatically groups strokes in the input vector sketch and then replaces each group by the best corresponding fitting curve—a procedure we call sketch consolidation. We then present a method that detects and resolves unintended gaps in a consolidated vector line drawing using learned local classifiers and global cues. Finally, we propose StripMaker, a consolidation method that jointly considers local perception cues from the first method and connectivities detected by the second method. We further integrate observations about temporal and contextual information present in drawing, resulting in a method with superior consolidation performance and potential for better user interactivity. Together, this work identifies important factors in humans’ perception of freehand sketches and provides automatic tools that narrow the gap between the raw freehand vector sketches directly created by artists and the requirements of downstream computational applications.

Bayesian modelling of DNA secondary structure kinetics : revisiting path space approximations and posterior inference in exponentially large state spaces
Lovrod, Jordan
DOI : 10.14288/1.0431312
URI : http://hdl.handle.net/2429/84372
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Anne Condon

Advancing variational inference via thermodynamic integration
Masrani, Vaden
DOI : 10.14288/1.0430543
URI : http://hdl.handle.net/2429/84173
Degree : Doctor of Philosophy - PhD
Graduation Date : 2023-05
Supervisor : Dr. Frank Wood

Variational inference (VI) is a popular method used within statistics and machine learning to approximate intractable probability distributions via optimization. Central to VI is the Evidence Lower Bound (ELBO), a variational objective function which lower bounds the log marginal likelihood, and can be used to jointly perform maximum likelihood parameter estimation and approximate posterior inference using stochastic gradient ascent. The core contribution of this thesis is the Thermodynamic Variational Objective (TVO), a novel variational objective derived from a key connection we make between variational inference and thermodynamic integration. The TVO both tightens and generalizes the ubiquitous ELBO, and empirically leads to improvements in model and inference network learning in both discrete and continuous deep generative models. Using a novel exponential family interpretation of the geometric mixture curve underlying the TVO, we characterize the divergence bound gap left by the TVO as a sum of KL divergences between adjacent distributions, with the forward and reverse KL’s corresponding to the lower and upper-bound TVO variants. To enable the TVO to be used in gradient- based optimization algorithms, we provide two computationally efficient score-function and doubly-reparameterized based gradient estimators, as well as two adaptive “schedulers” which choose the discretization locations of a one- dimensional Riemann integral approximation, a key hyperparameter in the TVO. Additionally, we show that the objective functions used in Variational Inference, Variational AutoEncoders, Wake-Sleep, Inference Compilation, and Rényi Divergence Variational Inference are all special cases of the TVO. Finally, we evaluate the TVO in two real-world settings - a stochastic control flow models with discrete latent variables, and multi-agent trajectory prediction with continuous latent variables built on top of a differentiable driving simulator - and find the TVO improves upon baseline objectives in both cases.

[no title]
Minns, Jocelyn
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Ian Mitchell

Insights from infinitely wide neural networks
Mohamadi, Mohamad Amin
DOI : 10.14288/1.0431593
URI : http://hdl.handle.net/2429/84542
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Danica Sutherland

Accelerating Bayesian Inference in Probabilistic Programming
Munk, Andreas
DOI : 10.14288/1.0435206
URI :http://hdl.handle.net/2429/85481
Degree : Doctor of Philosophy - PhD
Graduation Date : 2023-11
Supervisor : Dr. Frank Wood

This thesis focuses on the use of Bayesian inference and its practical application in real-world scenarios, such as in scientific stochastic simulators, via probabilistic programming. The execution of a probabilistic program (the reference program) is synonymous with probabilistic inference, and requires "only" the explicit denotation of random variables and their distributions. The inference procedure is carried out by a general-purpose inference backend (or engine) at runtime. The efficacy of many inference backends is a combination of (1) how well the particular engine can capture complex dependency structures between latent variables and (2) the speed at which the reference program runs---i.e. how fast its joint probability distribution can be calculated. Improvements to either attribute will result in more efficient inference. Furthermore, as these inference procedures---and Bayesian inference in general---require exact conditioning, it is not immediately obvious how to carry out inference when the conditional observations are associated with uncertainty. The main contributions of this thesis are the improvement of existing inference approaches in probabilistic programming by adding an attention mechanism to the inference back-end known as inference compilation and the extension of probabilistic programming to facilitate automated surrogate modeling. Additionally, this thesis make theoretical contributions to the problem of performing inference when observations are associated with uncertainty. In summary, this thesis aims to further advance the applicability of probabilistic programming in scientific simulators and Bayesian inference, with contributions focused on improving existing inference approaches, extending the functionality of probabilistic programming, and providing theoretical solutions to the challenges of uncertainty associated with observations. The results of this thesis have the potential to benefit researchers across many fields, ranging from physics to finance, by providing a more efficient and practical approach to simulator inversion and Bayesian inference.

How Teens Can Build Remote Social Connection through an Emotionally Supportive Robot Swarm
Reid, Elizabeth
DOI : 10.14288/1.0435681
URI : http://hdl.handle.net/2429/85749
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Karon Maclean

A differentially private network traffic shaping framework
Sabzi, Amir
DOI : 10.14288/1.0436913
URI : http://hdl.handle.net/2429/86052
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Aastha Mehta & Dr. Mathias Lécuyer

Automatically associating resources with tasks based on a software developer’s activity
Salomon, Marie
DOI : 10.14288/1.0435196
URI : http://hdl.handle.net/2429/85459
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Gail Murphy

Fostering Empathy through Intergenerational Storytelling in Embodied Virtual Reality
Shen, Chenxinran
DOI : 10.14288/1.0435683
URI : http://hdl.handle.net/2429/85761
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Joanna McGrenere & Dr. Dongwook Yoon

2GP: Two-Phase Graph Partitioner
Sinaee, Hadi
DOI : 10.14288/1.0437222
URI : http://hdl.handle.net/2429/86218
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Margo Seltzer

Using Dual Height Fields for the Parametrization of Signed Distance Fields 
Talero, Camilo
DOI : 10.14288/1.0437181
URI : http://hdl.handle.net/2429/86171
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Alla Sheffer

[no title]
Tiwary, Mayank
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Ivan Beschastnikh & Dr. Thomas Pasquier

Vertex-and-Edge Ordering for Faster Parallel Graph Processing
Trostanovsky, Alex
DOI : 10.14288/1.0437140
URI : http://hdl.handle.net/2429/86136
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Margo Seltzer

Sarwat: A Rule-Based Intrusion Detection System for Self-Driving Laboratories
Wattoo, Zainab Saeed
DOI : 10.14288/1.0435846 
URI : http://hdl.handle.net/2429/85889
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Margo Seltzer

Disentangling the Latent Space of 3D Human Body Meshes
Wu, Yuhao
DOI : 10.14288/1.0437307 
URI : http://hdl.handle.net/2429/86287
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Dinesh Pai & Dr. Chang Shu

Neural Fourier Filter Bank
Wu, Zhijie
DOI : 10.14288/1.0435529
URI : http://hdl.handle.net/2429/85566
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Kwang Moo Yi & Dr. Alla Sheffer

Variations on sparsifier constructions
Xiao, Ke Han
DOI : 10.14288/1.0432033
URI : http://hdl.handle.net/2429/84582
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Joel Friedman

Redex-Plus: A Metanotation for Programming Languages
Xu, Junfeng 
DOI : 10.14288/1.0435516
URI : http://hdl.handle.net/2429/85550
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. William Bowman

Explicit and implicit warping for accurate human pose estimation and low-latency neural renderin
Yu, Frank
DOI : 10.14288/1.0431314
URI : http://hdl.handle.net/2429/84370
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Helge Rhodin

Augmenting metadata tags in open data tables using schema matching in a pay-as-you-go fashion
Yu, Haoran
DOI : 10.14288/1.0428551
URI : http://hdl.handle.net/2429/83992
Degree : Master of Science - MSc
Graduation Date : 2023-05
Supervisor : Dr. Rachel Pottinger

Optimistic Thompson sampling: strategic exploration in bandits and reinforcement learning
Zhang, Tianyue H.
DOI : 10.14288/1.0435839
URI : http://hdl.handle.net/2429/85881
Degree : Master of Science - MSc
Graduation Date : 2023-11
Supervisor : Dr. Mark Schmidt