A multidisciplinary research team from the University of Oxford recently developed a GPU-accelerated limit order book (LOB) simulator called JAX-LOB, the first of its kind.
JAX is a tool for training high-performance machine learning systems developed by Google. In the context of a LOB simulator, it allows artificial intelligence models to train directly on financial data.
The Oxford research team created a novel method by which JAX could be used to run a LOB simulator using only GPUs. Traditionally, LOB sims are ran using computer processing units (CPUs). By running them directly on a GPU chain, where modern AI training occurs, AI models are able to skip several communication steps. According to the Oxford team’s pre-print research paper, this gives a speed increase of up to 7X.
LOB dynamics are among the most scientifically studied facets of finance. In the stock market, for example, LOBs allow full-time traders to maintain liquidity throughout daily sessions. And in the cryptocurrency world, LOBs are embraced at nearly every level by professional investors.
“Software like JAX-LOB is interesting as it seems like the exact sort of thing that a future powerful AI may use to conduct its own financial experiments.”