Discovery of Interpretable Physical Laws in Materials via Language-Model-Guided Symbolic Regression
This work provides a more efficient and effective way to discover interpretable physical laws from complex materials data, which is crucial for materials scientists seeking to understand and predict material behavior.
This paper addresses the challenge of discovering interpretable physical laws from high-dimensional materials data by using large language models to guide symbolic regression. The method successfully identified novel, accurate, and simpler formulas for properties like bulk modulus, band gap, and oxygen evolution reaction activity in perovskite materials, reducing the search space by approximately 10^5.
Discovering interpretable physical laws from high-dimensional data is a fundamental challenge in scientific research. Traditional methods, such as symbolic regression, often produce complex, unphysical formulas when searching a vast space of possible forms. We introduce a framework that guides the search process by leveraging the embedded scientific knowledge of large language models, enabling efficient identification of physical laws in the data. We validate our approach by modeling key properties of perovskite materials. Our method mitigates the combinatorial explosion commonly encountered in traditional symbolic regression, reducing the effective search space by a factor of approximately $10^5$. A set of novel formulas for bulk modulus, band gap, and oxygen evolution reaction activity are identified, which not only provide meaningful physical insights but also outperform previous formulas in accuracy and simplicity.