mathematics: 20 curated resources
mathematics
20 resources · page 1 of 1
American Mathematical Society: use of AI policy
by American Mathematical Society
American Mathematical Society policy on use of artificial intelligence in its journal publications.
DeepSeek-Prover
by DeepSeek
DeepSeek-Prover-V2 GitHub repo: Lean theorem proving models, ProverBench and download links.
Goedel-Prover
by Goedel-LM
Goedel-Prover-V2 GitHub repo: open-source language models for formal theorem proving in Lean.
Harmonic (Aristotle)
by Harmonic
Harmonic: company building Aristotle, an AI system for mathematical reasoning with formal proofs in Lean.
Lean theorem prover
by Lean FRO
Lean language site: open-source programming language and proof assistant for writing machine-checked mathematical proofs.
Lean community (mathlib)
Lean community site: Lean theorem prover, the mathlib library and documentation.
mathlib4
mathlib4 GitHub repo: community mathematics library for Lean 4.
miniF2F
by OpenAI
miniF2F GitHub repo from OpenAI: formal olympiad-level mathematics benchmark across several proof assistants.
The Well (Polymathic AI)
Polymathic AI's The Well: about 15 TB of physics simulation datasets for training and benchmarking ML surrogate models.
PutnamBench
by Trishul: Trustworthy Intelligent Systems @ UT Austin
PutnamBench GitHub repo: Putnam competition problems formalised in Lean, Isabelle and Coq for evaluating theorem-proving systems.
AI achieves silver-medal standard solving International Mathematical Olympiad problems (Google DeepMind)
Google DeepMind post on AlphaProof and AlphaGeometry 2 reaching silver-medal level at IMO 2024.
Mathematics for Machine Learning
by mml-book
Mathematics for Machine Learning book site by Deisenroth, Faisal and Ong, with free PDF on linear algebra, calculus and probability.
MIT 18.06 Linear Algebra (Strang)
MIT OpenCourseWare 18.06 Linear Algebra by Gilbert Strang, with video lectures, notes and problem sets.
AI achieves silver-medal standard solving International Mathematical Olympiad problems (AlphaProof)
Google DeepMind post on AlphaProof and AlphaGeometry 2 solving International Mathematical Olympiad problems at silver-medal level.
Datawhale pumpkin-book 南瓜书
by Datawhale
Datawhale Pumpkin Book: Chinese companion with step-by-step derivations of the formulas in Zhou Zhihua's Machine Learning textbook.
Equational Theories Project (Tao et al.)
by teorth
Terence Tao's Equational Theories Project: crowdsourced and automated-prover mapping of implications between equational laws.
Mathlib overview
Mathlib overview page mapping which areas of mathematics are formalised in the Lean mathlib library.
AlphaEvolve: A Gemini-powered coding agent for designing advanced algorithms
Google DeepMind post on AlphaEvolve, a Gemini-based coding agent that evolves algorithms for maths and computing problems.
Olympiad-Level Formal Mathematical Reasoning with Reinforcement Learning (AlphaProof)
Google Research publication on AlphaProof: olympiad-level formal mathematical reasoning with reinforcement learning.
MaxProof
by Jiacheng Chen et al. (Chinese University of Hong Kong)
arXiv paper (2026) introducing MaxProof, generative-verifier RL and test-time scaling for mathematical proofs.