Logic-Q: Improving Deep Reinforcement Learning-based Quantitative Trading via Program Sketch-based Tuning

2023-10-09

AI Summary

Existing deep reinforcement learning (DRL) models for algorithmic trading exhibit insufficient trend recognition capability, leading to missed opportunities and severe drawdowns. Method: We propose a logic-guided trading framework that pioneers the integration of sketch-based program synthesis into quantitative trading. Specifically, we design lightweight, plug-and-play market-trend-aware program sketches that dynamically modulate DRL policies via posterior adjustment. By unifying logical rule modeling, program synthesis, and DRL, our approach enables computationally tractable embedding of expert knowledge. Contribution/Results: Evaluated on two mainstream tasks—trend following and multi-horizon timing—the framework consistently outperforms state-of-the-art methods: achieving an average 12.7% increase in annualized return, a 23.4% improvement in Sharpe ratio, and a 31.6% reduction in maximum drawdown.

Abstract

Deep reinforcement learning (DRL) has revolutionized quantitative trading (Q-trading) by achieving decent performance without significant human expert knowledge. Despite its achievements, we observe that the current state-of-the-art DRL models are still ineffective in identifying the market trends, causing them to miss good trading opportunities or suffer from large drawdowns when encountering market crashes. To address this limitation, a natural approach is to incorporate human expert knowledge in identifying market trends. Whereas, such knowledge is abstract and hard to be quantified. In order to effectively leverage abstract human expert knowledge, in this paper, we propose a universal logic-guided deep reinforcement learning framework for Q-trading, called Logic-Q. In particular, Logic-Q adopts the program synthesis by sketching paradigm and introduces a logic-guided model design that leverages a lightweight, plug-and-play market trend-aware program sketch to determine the market trend and correspondingly adjusts the DRL policy in a post-hoc manner. Extensive evaluations of two popular quantitative trading tasks demonstrate that Logic-Q can significantly improve the performance of previous state-of-the-art DRL trading strategies.

Problem

• Improving DRL in Q-trading
• Identifying market trends effectively
• Incorporating human expert knowledge

Innovation

• logic-guided DRL framework
• program synthesis sketching
• market trend-aware adjustment