On the Universality of Online Mirror Descent
Nathan Srebro, Karthik Sridharan, Ambuj Tewari
TL;DR
This paper demonstrates that for a broad class of convex online learning problems, the Mirror Descent algorithm can consistently achieve near-optimal regret bounds, highlighting its universal applicability.
Contribution
It proves the universality of Mirror Descent in achieving optimal regret across various convex online learning scenarios.
Findings
Mirror Descent attains near-optimal regret in general convex online problems
The results establish the broad applicability of Mirror Descent
Theoretical guarantees for Mirror Descent's performance
Abstract
We show that for a general class of convex online learning problems, Mirror Descent can always achieve a (nearly) optimal regret guarantee.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdvanced Bandit Algorithms Research · Optimization and Search Problems · Machine Learning and Algorithms