Neuro-symbolic Meta Reinforcement Learning for Trading
This addresses trading efficiency for financial practitioners, but it is incremental as it builds on existing meta-RL and symbolic methods.
The paper tackled short-duration trading under uncertainty and concept-drift by combining meta reinforcement learning with symbolic pattern discovery, reporting that meta-RL outperforms vanilla RL and benefits from symbolic features.
We model short-duration (e.g. day) trading in financial markets as a sequential decision-making problem under uncertainty, with the added complication of continual concept-drift. We, therefore, employ meta reinforcement learning via the RL2 algorithm. It is also known that human traders often rely on frequently occurring symbolic patterns in price series. We employ logical program induction to discover symbolic patterns that occur frequently as well as recently, and explore whether using such features improves the performance of our meta reinforcement learning algorithm. We report experiments on real data indicating that meta-RL is better than vanilla RL and also benefits from learned symbolic features.