The unknown theorist: CMU's Andrej Risteski writes the math behind why deep learning works
bigaiguy · x · 2026-10-23
A long-form profile of Andrej Risteski, associate professor at CMU's Machine Learning Department and one of the least-known people shaping deep learning theory. PhD at Princeton under Sanjeev Arora, Norbert Wiener Fellow at MIT, he studies why generative models work, when out-of-distribution generalization happens, and the guarantees beneath self-supervised learning (e.g., "Provable Benefits of Score Matching"). He won an NSF CAREER Award ($500K over five years) and an Amazon Research Award, and his students/collaborators now work at OpenAI, Google, MIT, and Stanford. The thesis: the formal foundations of modern AI are being written by people the industry rarely talks about.
More from Companies & People
- IBM CEO: AI could let enterprises cut 490 of 500+ software vendors — victor_explore · 2026-10-23
- Google Discover tests blocking clicks on AI summary sources — lilyraynyc · 2026-10-23
- Andreessen: employees are secretly using AI and not telling management — rohanpaul_ai · 2026-10-23
- Every's frontier team and a tab-free AI browser built around one engineer's workflow — every · 2026-10-23
- Demis Hassabis's AGI timeline reportedly shrinks from 10 years to under 2 — TinfoilTricorn · 2026-10-23
- RIP Data Scientists: The Generative AI Data Scientist Is What Companies Now Want — mdancho84 · 2026-10-23