Is deep learning a useful tool for the pure mathematician?

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

Geordie Williamson This is my paper · ORCID

classification math.RTcs.LGmath.AGmath.CO

keywords deeplearningmathematicianpureaccountexpectinformalmight

verification ladder T0 review T1 audit T2 compute T3 formal

0 comments

read the original abstract

A personal and informal account of what a pure mathematician might expect when using tools from deep learning in their research.

Forward citations

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Neural Discovery in Mathematics: Do Machines Dream of Colored Planes?
cs.LG 2025-01 conditional novelty 6.0 of 10

A neural network relaxation of geometric coloring constraints produced new plane colorings, including an almost 5-coloring covering all but 3.74% of the plane, improving known bounds for Hadwiger-Nelson variants.
A Hybrid Framework for Healing Semigroups with Machine Learning
math.RA 2025-09 conditional novelty 5.0 of 10

A hybrid random-forest-plus-deterministic method heals corrupted finite semigroup tables, restoring associativity in 95% of small cases and 60% at n=10.
Explaining Deep Network Classification of Matrices: A Case Study on Monotonicity
cs.LG 2025-07 conditional novelty 5.0 of 10

For random 7x7 matrices with entries uniform in (-1,1), the ratio of the two lowest characteristic-polynomial coefficients, equal to 1/tr(A^{-1}) for monotone A, is empirically below 0.1755 for all 18,000 sampled mono...