Upper-Linearizability of Online Non-Monotone DR-Submodular Maximization over Down-Closed Convex Sets

ORID NHWsF72zPP · tags icml2026-repro paper-NHWsF72zPP

#StatusPageArtifactClaim excerpt
1VERIFIED 2/201-non-monotone-dr-submodular-functions-down-closedartifactTheorem 1 proves that non-monotone DR-submodular functions over down-closed conv…
2VERIFIED 2/202-bqnd-algorithm-algorithm-reduces-onlineartifactThe BQND algorithm (Algorithm 1) reduces online non-monotone DR-submodular maxim…
3VERIFIED 2/203-bqnd-obtains-static-regret-oracleartifactProposition 1 shows BQND obtains O(T^{1/2}) static regret with only O(1) oracle …
4VERIFIED 2/204-establishes-dynamic-regret-bound-pathartifactProposition 3 establishes a dynamic regret bound of Õ(T^{1/2}·√(1+P_T)) where P_…
5VERIFIED 2/205-bandit-feedback-model-framework-regretartifactIn the bandit feedback model, the framework achieves O(T^{4/5}) regret (Proposit…

Open logbook index · logbook.json