Research map / Reinforcement learning / Policy gradient and actor-critic
HyMTRL: A Hybrid Multi-Task Reinforcement Learning Framework via Phased Policy Evolution
Jinmin He, Kai Li, Xiaoyi Dong, Yifan Zang, Yuheng Jing, Yifan Zhang, Junliang Xing, Jian Cheng
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 |
|---|---|---|---|
| add HyMTRL | MOORE · MetaWorld-MT50 | success rate (%), all 50 tasks: 68.1 -> 86.2 | Improved |
| add HyMTRL | MTSAC · MetaWorld-MT50 | success rate (%), all 50 tasks: 48.4 -> 71.7 | Improved |
| add HyMTRL | SM · MetaWorld-MT50 | success rate (%), all 50 tasks: 55.6 -> 70.4 | Improved |
| add HyMTRL | MOORE · MetaWorld-MT50 | bottom-10 success rate (%): 0.8 -> 37.3 | Improved |
| add HyMTRL | MHSAC · MetaWorld-MT50 | bottom-10 success rate (%): 0.0 -> 0.0 | No clear effect |
| add HyMTRL | MHSAC · MetaWorld-MT10 | success rate (%): 62.3 -> 80.1 | Improved |
| add critic reset (critic only) (ablation) | HyMTRL MHSAC · MetaWorld-MT50 | success rate (%), all 50 tasks: 65.9 -> 72.6 | Improved |
| reset actor as well as critic (ablation) | HyMTRL MHSAC · MetaWorld-MT50 | success rate (%), all 50 tasks: 72.6 -> 67.9 | Hurt |
| add critic reset to baseline without HyMTRL (ablation) | MHSAC · MetaWorld-MT50 | success rate (%), all 50 tasks: 51.6 -> 57.1 | Improved |
| raise mastery threshold from 0.9 to 0.99 (ablation) | HyMTRL MHSAC · MetaWorld-MT50 | success rate (%), all 50 tasks: 72.6 -> 59.4 | Hurt |
| lower mastery threshold from 0.9 to 0.5 (ablation) | HyMTRL MHSAC · MetaWorld-MT50 | success rate (%), all 50 tasks: 72.6 -> 60.7 | Hurt |
| use critic reset interval 3e5 vs 1e5 (ablation) | HyMTRL MHSAC · MetaWorld-MT50 | success rate (%), all 50 tasks: 53.7 -> 72.6 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PaCo — PaCo: Parameter-Compositional Multi-task Reinforcement Learning (NeurIPS 2022) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- AMAGO-2: Breaking the Multi-Task Barrier in Meta-Reinforcement Learning with Transformers (NeurIPS 2024 · similarity 0.88)
- TOPPO: Rethinking PPO for Multi-Task Reinforcement Learning with Critic Balancing (NeurIPS 2026 · similarity 0.88)
- Hybrid Inverse Reinforcement Learning (ICML 2024 · similarity 0.87)
- QMP: Q-switch Mixture of Policies for Multi-Task Behavior Sharing (ICLR 2025 · similarity 0.87)
- Efficient Multi-task Reinforcement Learning with Cross-Task Policy Guidance (NeurIPS 2024 · similarity 0.87)
- Representation Learning Enables Scalable Multitask Deep Reinforcement Learning (NeurIPS 2026 · similarity 0.87)
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.
