Research map / Reinforcement learning / MDPs and planning
Anytime-Competitive Reinforcement Learning with Policy Prior
Jianyi Yang, Pengfei Li, Tongxin Li, Adam Wierman, Shaolei Ren
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Apply ACD safe-action projection to the learned policy | RL policy · Carbon-aware resource management | anytime-constraint violation rate: 27.5% -> zero violation | Improved |
| Apply ACD safe-action projection to the learned policy | RL policy · Carbon-aware resource management | anytime-constraint violation rate: 58.6% -> zero violation | Improved |
| Train the policy in the ACD-induced MDP rather than the original environment | RL+ACD · Carbon-aware resource management | regret | Improved |
| Use anytime cost constraints instead of expected cost constraints | Sustainable AI inference · Sustainable AI inference | regret | Hurt |
| Increase the constraint relaxation parameter λ | ACRL · Sustainable AI inference | regret | Improved |
| Increase λ or b to relax the anytime constraints | ACRL · Carbon-aware resource management | regret | Improved |
| Use anytime-safe ACD action selection instead of unconstrained reward optimization | ACRL · Carbon-aware resource management | anytime-constraint violation rate | Improved |
| Use a randomly initialized ML policy as the ACD input | Random+ACD · Carbon-aware resource management | regret: 61 to 64.51 | Hurt |
Closest papers by meaning
- Efficient Exploration in Average-Reward Constrained Reinforcement Learning: Achieving Near-Optimal Regret With Posterior Sampling (ICML 2024 · similarity 0.90)
- Learning General Parameterized Policies for Infinite Horizon Average Reward Constrained MDPs via Primal-Dual Policy Gradient Algorithm (NeurIPS 2024 · similarity 0.89)
- Achieving $\tilde{O}(1/\epsilon)$ Sample Complexity for Constrained Markov Decision Process (NeurIPS 2024 · similarity 0.88)
- Provably Efficient RL under Episode-Wise Safety in Constrained MDPs with Linear Function Approximation (NeurIPS 2025 · similarity 0.88)
- Global Convergence for Average Reward Constrained MDPs with Primal-Dual Actor Critic Algorithm (NeurIPS 2025 · similarity 0.88)
- ACPO: A Policy Optimization Algorithm for Average MDPs with Constraints (ICML 2024 · similarity 0.88)
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.
