Browsing Mathematics (Faculty of) by Supervisor "Poupart, Pascal"
Now showing items 1-19 of 19
-
Applications of Artificial Intelligence to the NHL Entry Draft
(University of Waterloo, 2019-01-31)This thesis investigates the application of various fields of artificial intelligence to the domain of sports management and analysis. The research in this thesis is primarily focussed on the entry draft for the National ... -
Bayesian Federated Learning in Predictive Space
(University of Waterloo, 2023-08-10)Federated Learning (FL) involves training a model over a dataset distributed among clients, with the constraint that each client's data is private. This paradigm is useful in settings where different entities own different ... -
Directly Learning Tractable Models for Sequential Inference and DecisionMaking
(University of Waterloo, 2016-06-21)Probabilistic graphical models such as Bayesian networks and Markov networks provide a general framework to represent multivariate distributions while exploiting conditional independence. Over the years, many approaches ... -
Emotion-Aware and Human-Like Autonomous Agents
(University of Waterloo, 2019-12-20)In human-computer interaction (HCI), one of the technological goals is to build human-like artificial agents that can think, decide and behave like humans during the interaction. A prime example is a dialogue system, where ... -
Entropy-based aggregate posterior alignment techniques for deterministic autoencoders and implications for adversarial examples
(University of Waterloo, 2020-08-27)We present results obtained in the context of generative neural models — specifically autoencoders — utilizing standard results from coding theory. The methods are fairly elementary in principle, yet, combined with the ... -
Generalization on Text-based Games using Structured Belief Representations
(University of Waterloo, 2020-12-23)Text-based games are complex, interactive simulations where a player is asked to process the text describing the underlying state of the world to issue textual commands for advancing in a game. Playing these games can be ... -
Incorporating Linear Dependencies into Graph Gaussian Processes
(University of Waterloo, 2023-08-28)Graph Gaussian processes are an important technique for learning unknown functions on graphs while quantifying uncertainty. These processes encode prior information by using kernels that reflect the structure of the graph, ... -
Learn Privacy-friendly Global Gaussian Processes in Federated Learning
(University of Waterloo, 2022-08-17)In the era of big data, Federated Learning (FL) has drawn great attention as it naturally operates on distributed computational resources without the need of data warehousing. Similar to Distributed Learning (DL), FL ... -
Likelihood-based Density Estimation using Deep Architectures
(University of Waterloo, 2019-12-20)Multivariate density estimation is a central problem in unsupervised machine learning that has been studied immensely in both statistics and machine learning. Several methods have thus been proposed for density estimation ... -
Linearizing Contextual Multi-Armed Bandit Problems with Latent Dynamics
(University of Waterloo, 2022-02-10)In many real-world applications of multi-armed bandit problems, both rewards and observed contexts are often influenced by confounding latent variables which evolve stochastically over time. While the observed contexts and ... -
Method of Moments in Approximate Bayesian Inference: From Theory to Practice
(University of Waterloo, 2021-07-12)With recent advances in approximate inference, Bayesian methods have proven successful in larger datasets and more complex models. The central problem in Bayesian inference is how to approximate intractable posteriors ... -
Model-Based Bayesian Sparse Sampling for Data Efficient Control
(University of Waterloo, 2019-06-24)In this work, we propose a novel Bayesian-inspired model-based policy search algorithm for data efficient control. In contrast to other model-based approaches, our algorithm makes use of approximate Gaussian processes in ... -
Multi-Resolution and Asymmetric Implementation of Attention in Transformers
(University of Waterloo, 2022-04-29)Transformers are the state-of-the-art for machine translation and grammar error correction. One of the most important components of transformers are the attention layers, but they require significant computational power. ... -
Naive Bayes Data Complexity and Characterization of Optima of the Unsupervised Expected Likelihood
(University of Waterloo, 2017-09-21)The naive Bayes model is a simple model that has been used for many decades, often as a baseline, for both supervised and unsupervised learning. With a latent class variable it is one of the simplest latent variable models, ... -
On the relationship between satisfiability and partially observable Markov decision processes
(University of Waterloo, 2018-09-26)Stochastic satisfiability (SSAT), Quantified Boolean Satisfiability (QBF) and decision-theoretic planning in finite horizon partially observable Markov decision processes (POMDPs) are all PSPACE-Complete problems. Since ... -
Online Bayesian Learning in Probabilistic Graphical Models using Moment Matching with Applications
(University of Waterloo, 2016-05-18)Probabilistic Graphical Models are often used to e fficiently encode uncertainty in real world problems as probability distributions. Bayesian learning allows us to compute a posterior distribution over the parameters of ... -
Parameter and Structure Learning Techniques for Sum Product Networks
(University of Waterloo, 2019-09-25)Probabilistic graphical models (PGMs) provide a general and flexible framework for reasoning about complex dependencies in noisy domains with many variables. Among the various types of PGMs, sum-product networks (SPNs) ... -
Prompt-tuning in Controlled Dialogue Generation
(University of Waterloo, 2022-12-22)Recent years have witnessed a prosperous development of dialogue response generation since the advent of Transformer. Fine-tuning pretrained language models for different downstream tasks has become the dominant paradigm ... -
Understanding Minimax Optimization in Modern Machine Learning
(University of Waterloo, 2021-07-21)Recent years has seen a surge of interest in building learning machines through adversarial training. One type of adversarial training is through a discriminator or an auxiliary classifier, such as Generative Adversarial ...