Research map / Reinforcement learning / Visual and representation RL
Coordinated Humanoid Robot Locomotion with Symmetry Equivariant Reinforcement Learning Policy
Buqing Nie, Yang Zhang, Rongjun Jin, Zhanxiang Cao, Huangxuan Lin, Xiaokang Yang, Yue Gao
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 strict symmetry equivariant actor and invariant critic | DreamWaQ · Unitree G1 velocity tracking (simulation) | TE-V (cm/s): 16.43 -> 9.85 | Improved |
| add strict symmetry equivariant actor and invariant critic | DreamWaQ-Regu · Unitree G1 velocity tracking (simulation) | TE-V (cm/s): 13.91 -> 9.85 | Improved |
| add strict symmetry equivariant actor and invariant critic | DreamWaQ · Unitree G1 velocity tracking (simulation) | Temp-S (1e-2 rad): 22.52 -> 7.86 | Improved |
| add strict symmetry equivariant actor and invariant critic | DreamWaQ · Unitree G1 velocity tracking (simulation) | Spat-S (1e-2 rad): 30.84 -> 0.00 | Improved |
| add soft symmetry regularization loss | DreamWaQ · Unitree G1 velocity tracking (simulation) | TE-V (cm/s): 16.43 -> 13.91 | Improved |
| add soft symmetry regularization loss | DreamWaQ · Unitree G1 velocity tracking (simulation) | Spat-S (1e-2 rad): 30.84 -> 8.18 | Improved |
| add symmetry-invariant critic (ablation) | SE-Policy actor only · Unitree G1 velocity tracking (simulation) | TE-V (cm/s): 11.06 -> 9.85 | Improved |
| add symmetry-invariant critic (ablation) | SE-Policy actor only · Unitree G1 velocity tracking (simulation) | Temp-S (1e-2 rad): 9.20 -> 7.86 | Improved |
| add symmetry-invariant critic (ablation) | SE-Policy actor only · Unitree G1 velocity tracking (simulation) | Spat-S (1e-2 rad): 0.00 -> 0.00 | No clear effect |
Closest papers by meaning
- Adversarial Locomotion and Motion Imitation for Humanoid Policy Learning (NeurIPS 2025 · similarity 0.88)
- Time Reversal Symmetry for Efficient Robotic Manipulations in Deep Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- HumanoidArena: Benchmarking Egocentric Hierarchical Whole-body Learning (NeurIPS 2026 · similarity 0.86)
- A Practical Guide for Incorporating Symmetry in Diffusion Policy (NeurIPS 2025 · similarity 0.86)
- Hierarchical World Models as Visual Whole-Body Humanoid Controllers (ICLR 2025 · similarity 0.86)
- ${\rm E}(3)$-Equivariant Actor-Critic Methods for Cooperative Multi-Agent Reinforcement Learning (ICML 2024 · similarity 0.85)
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.
