Publication Type

PhD Dissertation

Version

publishedVersion

Publication Date

5-2026

Abstract

My goal is to build autonomous systems that expand the reach of human capability in challenging domains such as undersea and space exploration, disaster response, and large-scale infrastructure. In everyday settings, these systems will increasingly appear in safety-critical applications such as autonomous driving, robotics, and industrial manufacturing. A central requirement for these systems is the ability to operate reliably under uncertainty, particularly when the environment behaves in unanticipated ways.

The robust handling of unforeseen environment dynamics is therefore a technical cornerstone of autonomous decision-making; Adversarial attacks provide a useful and principled lens through which to study this problem. Adversarial \textit{robustness}, then, is both a practical starting point and a meaningful objective in its own right, particularly for systems deployed in safety-critical settings.

In this thesis, I establish a new foundation for adversarial reinforcement learning, motivated by the comparative structure of regret. Existing approaches to robustness typically rely on adversarial training, which often fails to generalize to novel attacks, or worst-case optimization, which provides lower-bound guarantees but tends to produce overly conservative policies. To address these limitations, I propose a framework guided by three key principles. First, robustness should be evaluated at the trajectory level, ensuring stability over long sequences of decisions rather than individual actions. Second, robust agents must be designed with future adversaries in mind, rather than optimized against a fixed perturbation strategy. Finally, principled descriptions of the underlying problem structure are essential: methods that exploit the true structure of adversarial decision-making remain robust as applications evolve, while purely heuristic approaches often prove brittle.

Building on these principles, this dissertation develops scalable regret-based formulations of robustness, analyzes the structural role of partial observability in adversarial reinforcement learning, and demonstrates their effectiveness on both standard benchmarks and real-world applications.

Keywords

Robust Reinforcement Learning, Reinforcement Learning, Adversarial Reinforcement Learning, LLM Red Teaming

Degree Awarded

PhD in Computer Science

Discipline

Artificial Intelligence and Robotics

Supervisor(s)

VARAKANTHAM, Pradeep Reddy

First Page

1

Last Page

125

Publisher

Singapore Management University

City or Country

Singapore

Copyright Owner and License

Author

Download

Share

COinS