Skip to content
#reinforcement learning Book Open access

A Hierarchical Reinforcement Learning Framework for Formulaic Alpha Factor Generation and Adaptive Composition

Nov 2026 · Proceedings of the 35th ACM International Conference on Information and Knowledge Management · 0 citations · 43 references
Stock Market Forecasting Methods

Abstract

Alpha factor mining aims to automatically construct predictive signals from raw market features and is fundamental to stock return forecasting. Recent reinforcement learning (RL) approaches formulate formulaic alpha discovery as a sequential decision problem yet face three key challenges. First, factor generation suffers from delayed and sparse rewards: informative feedback is typically available only after a complete expression is generated, making credit assignment for intermediate decisions difficult and limiting exploration efficiency. Second, overly rapid policy convergence limits the diversity and novelty of discovered expressions, further exacerbating factor crowding. Third, static factor composition fails to adapt to non-stationary market conditions. To address these challenges, we propose AlphaHRL, a new hierarchical RL framework for joint factor generation and adaptive composition. At the low level, we introduce process-level reward shaping to provide dense supervision during expression generation, including validity and stability signals. Then, we perform a tree-structured decomposition and a memory-based comparison to reshape terminal rewards using novelty, redundancy, and marginal utility signals relative to a maintained factor pool. At the high level, we learn a dynamic allocation policy that reweighs factors based on recent information coefficient (IC) trajectories, enabling regime-aware composition of a mega-alpha. Extensive experiments demonstrate that AlphaHRL consistently improves both factor quality and portfolio performance relative to state-of-the-art baselines. Our code is published anonymously at https://github.com/ECNU-CILAB/AlphaHRL.

Read PDF

Similar papers

#machine learning Review Open access Oct 2014

Software development in startup companies: A systematic mapping study

The results indicate that software engineering work practices are chosen opportunistically, adapted and configured to provide value under the constrains imposed by the startup context.

Nicolò Paternoster, Carmine Giardino, M. Unterkalmsteiner et al. · 394 citations · ⚡54
#machine learning Review Open access Jun 2014

Why Early-Stage Software Startups Fail: A Behavioral Framework

This state-of-practice investigation was performed using a literature review followed by a multiple-case study approach and presents how inconsistency between managerial strategies and execution can lead to failure by means of a behavioral framework.

Carmine Giardino, Xiaofeng Wang, P. Abrahamsson · 175 citations · ⚡19
#machine learning Review Open access Oct 2016

“Failures” to be celebrated: an analysis of major pivots of software startups

This study conducts a case survey study based on the secondary data of the major pivots happened in 49 software startups, and demonstrates that customer need pivot is the most common among all pivot types.

Sohaib Shahid Bajwa, Xiaofeng Wang, Anh Nguyen-Duc et al. · 127 citations · ⚡15
#machine learning Review Open access May 2016

Key Challenges in Software Startups Across Life Cycle Stages

It is found that what perceived as biggest challenges by software startups do vary across different life cycle stages, even though its significance decreases when the learning focuses of the startups move from problem to solution and their products mature.

Xiaofeng Wang, Henry Edison, Sohaib Shahid Bajwa et al. · 62 citations · ⚡6

Related blog posts

Microsoft Research Blog Nov 16, 2026

TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient

AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.