Jayakumar Subramanian

Senior Research Scientist 2

San Jose

Jayakumar is a Senior Research Scientist 2 in Adobe Research, San Jose. He works in the areas of reinforcement learning in single and multi-agent systems, agent-based models, and game theory. His current focus is on agentic AI, multi-agent reinforcement learning, differentiable agent-based modeling, the use of transformers and large language models for reinforcement learning, and reinforcement learning with human feedback for transformer-based language models and diffusion models. He has done a dual degree (B.Tech and M.Tech) in Aerospace Engineering from IIT Bombay and a Ph.D. in Reinforcement Learning from McGill University.

Publications

Offline RL by Reward-Weighted Fine-Tuning for Conversation Optimization

Mukherjee, Subhojyoti., Lai, Viet., Addanki, Raghavendra., Rossi, Ryan., Yoon, David., Bui, Trung., Rao, Anup., Subramanian, Jayakumar., Kveton, Branislav. (Dec. 2, 2025)

39th Annual Conference on Neural Information Processing Systems (NeurIPS 2025)