Data-dependent Bounds with $T$-Optimal Best-of-Both-Worlds Guarantees in Multi-Armed Bandits using Stability-Penalty Matching

Documenti analoghi

• Replicability is Asymptotically Free in Multi-armed Bandits
  di: Komiyama, Junpei, et al.
  Pubblicazione: (2024)
• Suboptimal Performance of the Bayes Optimal Algorithm in Frequentist Best Arm Identification
  di: Komiyama, Junpei
  Pubblicazione: (2022)
• LC-Tsallis-INF: Generalized Best-of-Both-Worlds Linear Contextual Bandits
  di: Kato, Masahiro, et al.
  Pubblicazione: (2024)
• Near-optimal Per-Action Regret Bounds for Sleeping Bandits
  di: Nguyen, Quan, et al.
  Pubblicazione: (2024)
• Follow-the-Perturbed-Leader with Fréchet-type Tail Distributions: Optimality in Adversarial Bandits and Best-of-Both-Worlds
  di: Lee, Jongyeong, et al.
  Pubblicazione: (2024)