Publications
Selected research papers
GEM Workshop, ICLR 2024 & J. Chem. Inf. Model. (JCIM)
Interested in reinforcement learning and decision making. Using RL to solve problems, preferring real-world applications such as robotics but also new-age LLMs for fine-tuning and reasoning. Working and envisioning useful LLM-based agents that have access to tools such as simulators (or wet labs?), and compute to test and verify their hypotheses to automate research and generate novel knowledge. Agents need agency!
Selected research papers
Other work and projects
Always interested to join or help cool projects — reach out!
Core research area focused on developing novel deep RL algorithms, exploration strategies, and sample-efficient training methods.
MoreApplying reinforcement learning to real-world robotic systems, bridging the sim-to-real gap. Creator of BricksRL for accessible robotics research.
MoreTraining and fine-tuning large language models with SL, RL, and combinations thereof.
MoreAutomating complex workflows with AI agents. Applications include scientific research, engineering, and general work.
More
