Unified system improves equation discovery with adaptive solver choices
LegoSR: A Unified Framework for Self-improving Agent-Guided Scientific Discovery
Artificial IntelligenceMachine Learning
Summary
Finding equations that explain scientific data can be tricky because different tools use different methods and speak different “languages.” The authors created LegoSR, a system that brings together seven equation-finding tools under one roof, so their results can be compared and combined easily. LegoSR also includes an assistant that chooses the best tool for the problem based on ongoing results and past experience. This makes discovering equations from noisy data more reliable and efficient.
What this means in practice
- •For data science teams: Combine multiple symbolic equation solvers to improve accuracy in modeling complex scientific phenomena from noisy experimental data.
- •For engineering modelers: Use adaptive solver selection to discover or refine physical models when partial knowledge of equations is available.
Authors
Hao Mao, Xu T. Liu, Peng Zhao, Shuai Lu, Yuntian Chen
Abstract
Equation discovery solvers often use different data interfaces, equation representations and evaluation criteria, making their results difficult to compare. We present LegoSR, a unified framework for algebraic symbolic regression and differential equation discovery. LegoSR integrates seven solvers through a common equation representation and evaluation process, and supports searches constrained by partially known equations. Researchers can run solvers directly or use LegoSR Agent, which selects solvers and adapts searches based on solver knowledge, intermediate results and lessons from earlier tasks. On noisy ordinary and partial differential equation benchmarks, LegoSR Agent recovers nearly as many equations as the union of seven expert-tuned solvers and outperforms a generic agent with the same model, tools and budget by up to 29 percentage points. Three examples demonstrate basic use, adaptive solver selection and discovery guided by partial physical knowledge.