Learning Weakly Communicating Average-Reward CMDPs: Strong Duality and Improved Regret

Documenti analoghi

• Near-Optimal Primal-Dual Algorithm for Learning Linear Mixture CMDPs with Adversarial Rewards
  di: Yu, Kihyun, et al.
  Pubblicazione: (2026)
• Improved Regret Bound for Safe Reinforcement Learning via Tighter Cost Pessimism and Reward Optimism
  di: Yu, Kihyun, et al.
  Pubblicazione: (2024)
• Provably Efficient Infinite-Horizon Average-Reward Reinforcement Learning with Linear Function Approximation
  di: Chae, Woojin, et al.
  Pubblicazione: (2024)
• Learning Infinite-Horizon Average-Reward Linear Mixture MDPs of Bounded Span
  di: Chae, Woojin, et al.
  Pubblicazione: (2024)
• On Convergence of Average-Reward Q-Learning in Weakly Communicating Markov Decision Processes
  di: Wan, Yi, et al.
  Pubblicazione: (2024)