Elad Hazan recalls two decades of online convex optimization, from job hedge to AI-era math
HazanPrinceton · x · 2026-09-16
Princeton professor Elad Hazan published a personal recollection of two decades of online convex optimization (OCO). Highlights:
- He entered the field as a third-year PhD student as a hedge against failing to land a theory job, surrounded by prodigies like Ryan O'Donnell, Scott Aaronson, Irit Dinur and advisor Sanjeev Arora — and even organized a student finance study group as a backup plan.
- The field's roots trace to Zinkevich's OCO paper and Cover's universal portfolios, with anecdotes about collaborators like Satyen Kale.
- He closes with an observation: AI excels at exactly the math style STOC/FOCS rewards, but is weaker at inventing new frameworks.
A rare first-hand field history blending technical narrative with academic culture.
Related event: Princeton's Hazan Reflects on Two Decades of Online Convex Optimization(3 posts)→
More from AGI Musings
- ITIF: safer AI without slowing progress — the 10% extinction figure has no empirical basis — castrotech · 2026-09-17
- Castro argues braking AI also slows the research that makes it safer — castrotech · 2026-09-17
- Two kinds of platform: confusing app platforms with infra gets worse in the agent era — matt_slotnick · 2026-09-17
- Dan Selsam's AI risk statement called essential reading amid safety funding debate — JacquesThibs · 2026-09-17
- AI Made Your Workflow Faster, or Just Moved the Bottleneck? — Druss_ · 2026-09-17
- Chamath challenges Nadella: most people feel no magical AI productivity boost — rohanpaul_ai · 2026-09-17