AI Paper Atlas

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

Improved Regret Bounds for Bandit Combinatorial Optimization

Shinji Ito, Daisuke Hatano, Hanna Sumita, Kei Takemura, Takuro Fukunaga, Naonori Kakimura, Ken-Ichi Kawarabayashi

NeurIPS 2019 · Bandits and online learning › Online convex optimization and games · approach: Best-of-both-worlds bandit algorithms

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 binary {0,1} strongly-correlated losses to derive lower bound, removing the 1/sqrt(log T) factorLower bound for bandit combinatorial optimization · Bandit combinatorial optimization, no assumptionregret lower bound: Omega(sqrt(dk^3 T / log T)) -> Omega(sqrt(dk^3 T))Improved
Restrict losses from continuous [0,1]^d to binary {0,1}^dLower-bound construction · Bandit combinatorial optimizationhardness (regret lower bound)No clear effect
Apply the new lower bound to multitask bandit, bandit ranking and multiple-play banditSpecial cases · Bandit ranking with full permutationsregret lower bound tightnessImproved
Correlated versus independent losses across armsProblem hardness · Bandit combinatorial optimizationregret lower bound: Omega(sqrt(dk^2 T)) -> Omega(sqrt(dk^3 T))Improved

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