Research map / Reinforcement learning / MDPs and planning
Offline Actor-Critic for Average Reward MDPs
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a pessimistic actor-critic algorithm for offline policy optimization in infinite-horizon average-reward MDPs with large state spaces, using linear function approximation for the critic.
Problem
Offline policy optimization for infinite-horizon average-reward MDPs with large or infinite state spaces, where data coverage is limited and naive value estimates are overoptimistic.
Approach
A critic that computes a pessimistic estimate of the average reward under the current policy, using a computationally efficient linear function class, combined with an actor update (abstract truncated, so details beyond this are not stated).
Closest papers by meaning
- Regret Analysis of Average-Reward Unichain MDPs via an Actor-Critic Approach (NeurIPS 2025 · similarity 0.92)
- Global Convergence for Average Reward Constrained MDPs with Primal-Dual Actor Critic Algorithm (NeurIPS 2025 · similarity 0.92)
- A Sharper Global Convergence Analysis for Average Reward Reinforcement Learning via an Actor-Critic Approach (ICML 2025 · similarity 0.92)
- Regret Analysis of Policy Gradient Algorithm for Infinite Horizon Average Reward Markov Decision Processes (AAAI 2024 · similarity 0.91)
- Optimal Single-Policy Sample Complexity and Transient Coverage for Average-Reward Offline RL (NeurIPS 2025 · similarity 0.90)
- Learning General Parameterized Policies for Infinite Horizon Average Reward Constrained MDPs via Primal-Dual Policy Gradient Algorithm (NeurIPS 2024 · similarity 0.90)
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.