Roman Belaire

Ph.D. candidate at Singapore Management University; I'm advised by Pradeep Varakantham, affiliated with the CARE.ai lab. My research concerns adversarial robustness in reinforcement learning, and I also have interests in hierarchical RL, generalization, and AI safety. Lately I've found interest in applications of graph theory and information theory to these problems, as well as game theory (the obvious one). I enjoy exercising, playing games, eating, and being in the ocean.

Research

  • Regret-Based Defense in Adversarial Reinforcement Learning (link to demo) AAMAS 2024 (Arxiv)

  • By optimizing a novel form of regret, we train RL agents that are more robust than previous robustly trained value-optimizing agents. Our regret notion, CCER, provides a scalable, transferrable way to compute adversarial cumulative regret for actions across time steps.
  • Probabilistic Perspectives on Error Minimization in Adversarial Reinforcement Learning (under review)Preprint

  • We progress the formulation of observation-adversarial RL by recognizing its true structure—a POMDP. Leveraging this fact, our proposed methods achieve SOTA performance across all adversarial RL benchmarks.
  • Automated Benchmarking to Red-Team Large Language Models Work in progress

  • In cooperation with Singapore's Infocomm and Media Development Authority (IMDA), we are developing novel ways to benchmark production LLMs in an automated fashion, using RL as a baseline framework to search the perturbation space.

Education

Singapore Management University Singapore

Ph.D. in Computer Science May 2025
Presidential Doctoral Fellowship recipient, 2024

California State University, Fullerton Fullerton, California

Bachelor of Science in Computer Science May 2020

Previous Engineering Background

  • Associate Software Engineer:Toshiba America Business Solutions May 2021-August 2021
    Lake Forest, California

  • Machine Learning Engineer: CSUF January 2019-September 2019
    Fullerton, California

  • Software Engineer Intern: Collins Aerospace Summer 2019
    Cedar Rapids, Iowa

  • AI Curriculum Designer:Bytes&Bots September 2018-May 2019
    Irvine, California

What else do I do?

I love my work, however my true passion is the ocean. I try to schedule 1-2 dive trips per year, and I have aspirations to take and share some high-quality underwater photographs. Until then, here are a few favorites that I've taken:

Get in touch

If you want to collaborate, have a burning question, or just want to say hi, you can contact me through the links below.