srush Derives MNIST from Peano Axioms in 3000 Lines of Lean

Sasha Rush (srush) released 'Deep Learning Speedrun,' formally deriving the math needed to train MNIST from Peano axioms in about 3000 lines of Lean, answering his own question about the minimal mathematics required for deep learning.

2026-10-29 ~ 2026-10-29 · 2 related posts

• srush derives MNIST in 3,000 lines of Lean, minimal DL math is half a semester — srush_nlp · 2026-10-29
• Sasha Rush's Deep Learning Speedrun shows the minimal math to train MNIST — srush_nlp · 2026-10-29