AI Paper Atlas

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

Online Convex Optimization with Continuous Switching Constraint

Guanghui Wang, Yuanyu Wan, Tianbao Yang, Lijun Zhang

NeurIPS 2021 · Bandits and online learning › Online convex optimization and games · approach: Constrained and unconstrained OCO algorithms · 2 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
Tune the constant OGD step size to the switching budgetOGD on OCO-CSC with convex losses · OCO-CSC, convex; S∈[D√T,DT]regret upper boundImproved
Tune the constant OGD step size to the switching budgetOGD on OCO-CSC with convex losses · OCO-CSC, convex; S∈[D,D√T)regret upper boundImproved
Use a denominator offset c in the time-varying OGD step sizeOGD on OCO-CSC with strongly convex losses · OCO-CSC, strongly convex; S∈[\frac{2G}{\lambda}\log(T+1),DT]regret upper boundImproved
Set the denominator offset c according to the switching budgetOGD on OCO-CSC with strongly convex losses · OCO-CSC, strongly convex; S∈[0,\frac{2G}{\lambda}\log(T+1))regret upper boundImproved
Use classical OGD with a budget-calibrated step size instead of mini-batch OGDOCO-CSC with convex losses · OCO-CSC, convexregret upper boundImproved
Adapt adversary epochs to cumulative switching costLower-bound construction for OCO-CSC · OCO-CSC, convexregret lower boundImproved

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