Introducing JAX-LOB: A GPU-Accelerated limit order book simulator to unlock large scale reinforcement learning for trading for after-work quants
Frey, Li, Nagy, Sapora, Lu, Zohren, Foerster and Calinescu (2023) implement JAX-LOB: thousands of fixed-size LOBs in parallel on GPU, wrapped in gymnax for recurrent PPO execution. ~7× training vs CPU LOB. Plain-language intro. Not investment advice.