Regime-Adaptive Trajectory Mutation
A regime-adaptive trajectory mutation framework that classifies market states via Hidden Markov Models to condition LLM structural edits during financial factor discovery.
Source paper
Jun Han, Shuo Zhang, Wei Li, Zhi Yang, Yifan Dong, Tu Hu, Jialuo Yuan, Xiaomin Yu, Yumo Zhu, Fangqi Lou, Xin Guo, Zhaowei Liu, Tianyi Jiang, Ruichuan An, Jingping Liu, Biao Wu, Rongze Chen, Kunyi Wang, Yifan Wang, Sen Hu, Xinbing Kong, Liwen Zhang, Ronghao Chen, Huacan Wang
Generated by AI. The paper’s authors did not write these.
A regime-adaptive trajectory mutation framework that classifies market states via Hidden Markov Models to condition LLM structural edits during financial factor discovery.
A method that incorporates a differentiable turnover proxy directly into the reward function and generation prompts to penalize high-turnover trading signals natively.
A multi-objective evolutionary algorithm that balances IC and maximum drawdown by selecting parent trajectories from the Pareto Frontier.
This paper introduces QuantaAlpha, an evolutionary framework for automating financial alpha factor mining using Large Language Models (LLMs). By treating the mining process as a 'trajectory' of decisions—from hypothesis generation to code implementation—the system employs evolutionary operators like Mutation and Crossover to iteratively refine factors, ensuring semantic consistency and robustness against market noise.
The framework utilizes an agentic workflow where LLMs generate factor hypotheses and executable code based on a symbolic operator library. These factors are evolved via a genetic algorithm: 'Mutation' uses self-reflection to repair specific suboptimal steps in a trajectory, while 'Crossover' recombines successful segments from different parent trajectories. The process is constrained by semantic verification and complexity regularization.