Machine-Learning Mathematical Structures
This work provides a general review for researchers interested in using machine learning to formulate conjectures, improve computational methods, and explore mathematical structures, but it is incremental as it synthesizes existing experiments.
The paper reviews recent experiments that apply supervised machine learning to labeled mathematical data from various fields to extract structure, achieving comparative accuracy results across different problems.
We review, for a general audience, a variety of recent experiments on extracting structure from machine-learning mathematical data that have been compiled over the years. Focusing on supervised machine-learning on labeled data from different fields ranging from geometry to representation theory, from combinatorics to number theory, we present a comparative study of the accuracies on different problems. The paradigm should be useful for conjecture formulation, finding more efficient methods of computation, as well as probing into certain hierarchy of structures in mathematics.