A custom MARL (multi-agent reinforcement learning) environment where multiple agents trade against one another (self-play) in a zero-sum continuous double auction. Ray [RLlib] is used for training.
lstm quantitative-finance ray limit-order-book quantitative-trading financial-engineering market-microstructure zero-sum high-frequency-trading gym-environment ppo self-play double-auction multi-agent-reinforcement-learning rllib marl n-player zero-sum-games
-
Updated
Sep 19, 2026 - Python