Multi-Agent Reinforcement Learning for Financial AI - NVIDIA

Project overview

Financial systems are complex, adaptive, and multi-agent by nature. Markets are shaped by interacting actors: traders, institutions, algorithms, and AI systems themselves. Our research explores how multi-agent reinforcement learning can model these dynamics and help build the next generation of financial AI systems capable of learning, adapting, and coordinating in complex environments.

This research is supported by the NVIDIA Academic Grant Program.

Project Team

• Prof. Sophia Ananiadou
• Mr. Jimin Huang
• Dr. Kailai Yang