Advancing mathematics by guiding human intuition with AI

TLDR

Mathematics relies on discovering patterns and proving conjectures, and since the 1960s computers have aided this process, notably in the Birch and Swinnerton‑Dyer conjecture. The study demonstrates how machine learning can help mathematicians discover new conjectures and theorems by identifying patterns, attributing them, and guiding intuition. The authors outline a machine‑learning‑guided framework that identifies patterns between mathematical objects, attributes them, and uses these insights to guide intuition, then apply it to diverse pure‑mathematics problems—yielding new connections between knot theory and a candidate algorithm for symmetric groups. The work demonstrates that machine learning can uncover new conjectures and theorems, producing significant advances such as a new knot‑theory connection and a symmetric‑group algorithm, and suggests a fruitful collaboration model between mathematics and AI.

Abstract

The practice of mathematics involves discovering patterns and using these to formulate and prove conjectures, resulting in theorems. Since the 1960s, mathematicians have used computers to assist in the discovery of patterns and formulation of conjectures1, most famously in the Birch and Swinnerton-Dyer conjecture2, a Millennium Prize Problem3. Here we provide examples of new fundamental results in pure mathematics that have been discovered with the assistance of machine learning-demonstrating a method by which machine learning can aid mathematicians in discovering new conjectures and theorems. We propose a process of using machine learning to discover potential patterns and relations between mathematical objects, understanding them with attribution techniques and using these observations to guide intuition and propose conjectures. We outline this machine-learning-guided framework and demonstrate its successful application to current research questions in distinct areas of pure mathematics, in each case showing how it led to meaningful mathematical contributions on important open problems: a new connection between the algebraic and geometric structure of knots, and a candidate algorithm predicted by the combinatorial invariance conjecture for symmetric groups4. Our work may serve as a model for collaboration between the fields of mathematics and artificial intelligence (AI) that can achieve surprising results by leveraging the respective strengths of mathematicians and machine learning.