AI Paper Atlas

Research map / Bandits and online learning / Online convex optimization and games

Beyond Time-Average Convergence: Near-Optimal Uncoupled Online Learning via Clairvoyant Multiplicative Weights Update

Georgios Piliouras, Ryann Sim, EFSTRATIOS PANTELEIMON SKOULAKIS

NeurIPS 2022 · Bandits and online learning › Online convex optimization and games · approach: Last-iterate convergence in games · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use CMWU with a \log T-sparse average instead of OMWU with a time-averageUncoupled online learning in general-sum games · General-sum gamesCCE convergence rate: O\left({nV\log m\log^4 T}/{T}\right) -> O\left(nV \log m \log T /T\right)Improved
Use clairvoyant MWU updates that evaluate payoffs at the next strategy profileCMWU · General gamesregret: R_i(T) \leq \log |S_i| / \eta_iImproved
Choose step-size \eta<\frac{1}{(n-1)V}CMWU fixed-point update · General gamesfixed-point computation: \eta V (n-1)Improved
Use CMWU dynamics with \eta = 1/2nV and k =\lceil \log T \rceilCMWU dynamics · General gamesregret on sparse subsequence: -12nV\log mImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy