Princeton's Elad Hazan reflects on two decades of online convex optimization
HazanPrinceton · x · 2026-09-16
Princeton professor Elad Hazan published a long personal recollection of roughly two decades of online convex optimization (OCO).
- It began in his third year of grad school, when Zinkevich's OCO paper appeared and he read Cover's universal portfolios work with Satyen Kale and others
- He admits that, fearing he couldn't compete for a theory job alongside geniuses like Scott Aaronson, Irit Dinur, and Ryan O'Donnell, he even organized a student group to learn finance as a career hedge — they're all close friends today
- He offers a striking observation: AI excels exactly at what STOC/FOCS rewards, but is less good at inventing new frameworks
- The essay weaves in anecdotes about collaborators and academic friendships
A deep, substantive piece from a leading figure in the field, part research history and part memoir.
Related event: Princeton's Hazan Reflects on Two Decades of Online Convex Optimization(3 posts)→
More from Research
- OmniHarness: symbolic policy learning boosts generalizable visual generation — Xu Xu · 2026-09-17
- TokenRhythm Launches NeoHorse-1: 4B/9B Models Post-Trained on Agent Execution Traces — rohanpaul_ai · 2026-09-17
- Paper2Agent's auto-generated paper MCP beats paper + code repo — james_y_zou · 2026-09-17
- Cohere Labs Releases Open Book 'World Models from Scratch' with Live Sessions — Cohere_Labs · 2026-09-17
- Million-dollar AI swarms: progress will come from building 'unit tests' outside the models — mentalgeorge · 2026-09-17
- PyTorchCon to feature federated learning glaucoma study spanning 9 datasets in 7 countries — PyTorch · 2026-09-17