Search

Fall 2020

Chan, Alan

Policy gradient methods typically estimate both explicit policy and value functions. The long-extant view of policy gradient methods as approximate policy iteration---alternating between policy evaluation and policy improvement by greedification---is a helpful framework to elucidate algorithmic...

Fall 2022

Neumann,Samuel

Actor-Critics are a popular class of algorithms for control. Their ability to learn complex behaviours in continuous-action environments make them directly applicable to many real-world scenarios. These algorithms are composed of two parts - a critic and an actor. The critic learns to critique...

Spring 2022

Samani, Abolfazl

The concept of state is fundamental to a reinforcement learning agent. The state is the input to the agent's action-selection policy, value functions, and environmental model. A reinforcement learning agent interacts with the environment by performing actions and receiving observations, resulting...

Spring 2024

Young, Kenneth J.

In this dissertation, I investigate how we can exploit generic problem structure to make reinforcement learning algorithms more efficient. Generic problem structure means basic structure that exists in a wide range of problems (e.g., an action taken in the present does not influence the past), as...

Spring 2023

Wang, Zeyi

The intent of this thesis is to develop a high-performance open-source system that plans with a learned model and to understand the algorithm through extensive analysis. We formulate the problem of maximizing accumulated rewards in Markov Decision Processes, and we frame playing games as such...

Fall 2023

Ray, Debraj

Real-time strategy games require players to respond to short-term challenges (micromanagement) and long-term objectives (macromanagement) simultaneously to win. However, many players excel at one of these skills but not both. This research studies whether the burden of micromanagement can be...

Spring 2023

Stephens, Connor J

Many practical problems in fields ranging from online advertising to genomics can be framed as the task of selecting the best option from among several choices, based on a limited number of noisy evaluations of the quality of each choice. Pure exploration in multi-armed bandits is an...

Fall 2023

Pramanik, Subhojeet

The transformer architecture is effective in processing sequential data, both because of its ability to leverage parallelism, and because of its self-attention mechanism capable of capturing long-range dependencies. However, the self-attention mechanism is slow for streaming data, that is when...

Spring 2024

Naik, Abhishek

This dissertation develops simple and practical learning algorithms from first principles for long-lived agents. Formally, the algorithms are developed within the reinforcement learning framework for continuing (non-episodic) problems, in which the agent-environment interaction goes on ad...

Fall 2020

Sherstan, Craig

Research in artificial general intelligence aims to create agents that can learn from their own experience to solve arbitrary tasks in complex and dynamic settings. To do so effectively and efficiently, such an agent must be able to predict how its environment will change both dependently and...