List your company
ResearchFormerly WizWand

Regret Minimization on Episodic Tabular MDPs Stochastic Regime with Adversarial Corruption

2Regret Upper BoundDann et al. (2023, Theorem 4.3)
1.922.4633.54Jun 30, 2026
Updated 4mo ago

Evaluation results

MethodLinks
Dann et al. (2023, Theorem 4.3)Unknown transition=false, Data-dependent=FO, Bandit feedback=true2026.062
Li et al. (2026)Unknown transition=false, Data-dependent=true, Bandit feedback=true2026.062
This work (Theorem 4.1)Unknown transition=true, Data-dependent=true, Bandit feedback=false2026.062
This work (Theorem 4.2)Unknown transition=true, Data-dependent=true, Bandit feedback=true2026.062
Dann et al. (2023, Theorem 4.2)Unknown transition=false, Data-dependent=true, Bandit feedback=true2026.064