$\text{Alpha}^2$: Discovering Logical Formulaic Alphas using Deep Reinforcement Learning

TL;DR

This work proposes a novel framework for alpha discovery using DRL by formulating the alpha discovery process as program construction, and demonstrates its capability to identify a diverse set of logical and effective alphas, which significantly improves the performance of the final trading strategy.

Abstract

Alphas are pivotal in providing signals for quantitative trading. The industry highly values the discovery of formulaic alphas for their interpretability and ease of analysis, compared with the expressive yet overfitting-prone black-box alphas. In this work, we focus on discovering formulaic alphas. Prior studies on automatically generating a collection of formulaic alphas were mostly based on genetic programming (GP), which is known to suffer from the problems of being sensitive to the initial population, converting to local optima, and slow computation speed. Recent efforts employing deep reinforcement learning (DRL) for alpha discovery have not fully addressed key practical considerations such as alpha correlations and validity, which are crucial for their effectiveness. In this work, we propose a novel framework for alpha discovery using DRL by formulating the alpha discovery process as program construction. Our agent, $\text{Alpha}^2$, assembles an alpha program optimized for an evaluation metric. A search algorithm guided by DRL navigates through the search space based on value estimates for potential alpha outcomes. The evaluation metric encourages both the performance and the diversity of alphas for a better final trading strategy. Our formulation of searching alphas also brings the advantage of pre-calculation dimensional analysis, ensuring the logical soundness of alphas, and pruning the vast search space to a large extent. Empirical experiments on real-world stock markets demonstrates $\text{Alpha}^2$'s capability to identify a diverse set of logical and effective alphas, which significantly improves the performance of the final trading strategy. The code of our method is available at https://github.com/x35f/alpha2.

Figures (4)

• Figure 1: Expression tree
• Figure 2: Example of pruning the search space according to dimensional consistency. The left part shows an ongoing search process of MCTS. The right part shows the expression trees of the corresponding alpha expressions. For an alpha expression $close - open$, which is of dimension $currency$, consider adding $high$ or $volume$ to the expression. Adding $high$ is allowed since it is of the same dimension $currency$, while adding $volume$ is forbidden because it is of another dimension. The nodes marked in green illustrates the addition of $high$, and the nodes marked in red illustrates the addition of $volume$.
• Figure 3: The pipeline for the generation of strategy
• Figure 4: Backtest result on CSI300. The value of the y axis represents the cumulative reward.