Research map / Label-efficient and robust learning / Domain adaptation and generalization
Curriculum Reinforcement Learning using Optimal Transport via Gradual Domain Adaptation
Peide Huang, Mengdi Xu, Jiacheng Zhu, Laixi Shi, Fei Fang, Ding Zhao
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 |
|---|---|---|---|
| use smaller delta alpha (0.2, 0.1, 0.05) (ablation) | GRADIENT · Maze | time-to-threshold | No clear effect |
| use GRADIENT curriculum instead of baseline curricula | GRADIENT (PPO) · Maze | time-to-threshold | Improved |
| uniform domain randomization instead of no curriculum | PPO learner · Maze | learning efficiency | Improved |
| uniform domain randomization instead of linear interpolation curriculum | PPO learner · Maze | learning efficiency | Improved |
| use GRADIENT or SPDL (target-aware) instead of other baselines | SAC learner · PointMass | time-to-threshold | Improved |
| use GRADIENT instead of SPDL | SAC learner · PointMass | time-to-threshold | Improved |
| use GRADIENT instead of SPDL on non-Gaussian target | SAC learner · FetchPush | time-to-threshold and asymptotic performance | Improved |
| use target-aware methods (SPDL, GRADIENT) instead of ALP-GMM and GoalGAN | SAC learner · FetchPush | performance | Improved |
| use domain randomization instead of ALP-GMM and GoalGAN | SAC learner · FetchPush | asymptotic performance | Improved |
| use learned distance embedding with episodic-reward surrogate metric instead of l2 distance (ablation) | GRADIENT · U-Maze | interpolation quality | Improved |
Closest papers by meaning
- Curriculum Reinforcement Learning via Constrained Optimal Transport (ICML 2022 · similarity 0.92)
- On the Benefit of Optimal Transport for Curriculum Reinforcement Learning (TPAMI 2024 · similarity 0.91)
- Diffusion-based Curriculum Reinforcement Learning (NeurIPS 2024 · similarity 0.89)
- Acquiring Diverse Skills using Curriculum Reinforcement Learning with Mixture of Experts (ICML 2024 · similarity 0.87)
- In-Context Policy Adaptation via Cross-Domain Skill Diffusion (AAAI 2025 · similarity 0.86)
- Curriculum Design for Trajectory-Constrained Agent: Compressing Chain-of-Thought Tokens in LLMs (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.
