Research map / Reinforcement learning / Policy gradient and actor-critic
Maximum Entropy-Regularized Multi-Goal Reinforcement Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Add MEP goal-entropy prioritization to DDPG | DDPG · FetchPush | success rate: 99.90% -> 99.96% | Helped |
| Add MEP goal-entropy prioritization to DDPG | DDPG · FetchPickAndPlace | success rate: 39.34% -> 76.02% | Helped |
| Add MEP goal-entropy prioritization to DDPG | DDPG · FetchSlide | success rate: 75.67% -> 76.77% | Helped |
| Add MEP goal-entropy prioritization to DDPG+HER | DDPG+HER · HandManipulateEgg | success rate: 76.19% -> 81.30% | Helped |
| Add MEP goal-entropy prioritization to DDPG+HER | DDPG+HER · HandManipulateBlock | success rate: 20.32% -> 25.00% | Helped |
| Add MEP goal-entropy prioritization to DDPG+HER | DDPG+HER · HandManipulatePen | success rate: 27.28% -> 31.88% | Helped |
| Prioritize transitions by TD-error using PER | DDPG · FetchSlide | success rate: 75.67% -> 66.33% | Hurt |
| Prioritize transitions by TD-error using PER | DDPG+HER · HandManipulateEgg and HandManipulateBlock | success rate: 76.19% -> 75.46%; 20.32% -> 18.95% | Hurt |
| Use MEP prioritization instead of uniform replay | DDPG and DDPG+HER · Six robotic environments | sample efficiency: 1.95 | Helped |
| Use MEP prioritization rather than PER | DDPG and DDPG+HER · Six robotic environments | training time: 4 times faster | Helped |
| Prioritize under-represented achieved-goal trajectories | MEP · Achieved goal distribution during training | goal entropy | Helped |
Closest papers by meaning
- Towards Principled Unsupervised Multi-Agent Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- Improving the Continuity of Goal-Achievement Ability via Policy Self-Regularization for Goal-Conditioned Reinforcement Learning (ICML 2025 · similarity 0.87)
- Active Fine-Tuning of Multi-Task Policies (ICML 2025 · similarity 0.87)
- The Max-Min Formulation of Multi-Objective Reinforcement Learning: From Theory to a Model-Free Algorithm (ICML 2024 · similarity 0.86)
- Convergence Theorems for Entropy-Regularized and Distributional Reinforcement Learning (NeurIPS 2025 · similarity 0.86)
- Multi-Objective Reinforcement Learning with Max-Min Criterion: A Game-Theoretic Approach (NeurIPS 2025 · similarity 0.86)
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.