GoAnt improves trading signals with diverse multi-agent search
GoAnt: Quality-Diversity Multi-Agent Search for Alpha Factor Discovery in Market Microstructure Data
Artificial Intelligence
Summary
Finding strong trading signals from complex market data is tough because many techniques end up repeating similar ideas or fail when trading costs are included. The authors created GoAnt, a method that uses different types of agents working independently but sharing information to find diverse, high-quality trading signals. GoAnt organizes and remembers the best signals in a smart way to avoid duplicates and focuses its efforts efficiently. Tested on real Chinese market data from 2023 to 2026, GoAnt delivered significantly better results than previous methods.
What this means in practice
- •For quantitative trading teams: Generate a wider variety of robust trading signals from detailed market data to improve trading strategies.
- •For financial data engineers: Deploy multi-agent search frameworks that efficiently allocate computational resources to discover diverse market predictors in complex datasets.
Authors
Stella Zhao, Tommy Sha
Abstract
Automated alpha factor discovery searches symbolic trading signals from price-volume panels and order-book data under a fixed evaluation budget. Existing single- and multi-agent program-search systems can overfit predictive proxies that fail after execution costs and repeatedly explore redundant factor families, limiting execution robustness and behavioral diversity. We introduce GoAnt, a quality-diversity multi-agent search framework that combines non-communicating Explorer, Exploiter and Connector workers with a shared adaptive Mental Map and a compact Queen dispatcher. The Mental Map organizes candidates by leakage-free execution profiles and retains one elite per niche, while the Queen reallocates the evaluation budget from explicit search-state summaries. We also define a map-independent effective-yield protocol that counts high-quality, mutually nonredundant factors directly from each method's evaluation records, giving archive-based and map-free systems the same ruler. On real A-share microstructure data spanning 2023--2026, GoAnt reaches quality-weighted yields of 41.8 and 47.6 in price-volume and order-book settings, improving the strongest baseline by 57% and 97% under matched budgets. Its locked populations retain 0.64 and 0.67 of in-sample quality out of sample, compared with 0.61 and 0.63 for a static map.