Research map / Reinforcement learning / Visual and representation RL
Grounded Action Transformation for Robot Learning in Simulation
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 |
|---|---|---|---|
| ground simulator with action transformation (GAT) vs no grounding (ablation) | CMA-ES in SimSpark · SimSpark to Gazebo (Gazebo as real world) | % improvement: 11.094 -> 22.48 | Helped |
| GAT vs Noise-Envelope (ablation) | CMA-ES in SimSpark · SimSpark to Gazebo (Gazebo as real world) | % improvement: 18.93 -> 22.48 | Helped |
| add Gaussian action noise envelope to ungrounded simulator (ablation) | CMA-ES in SimSpark · SimSpark to Gazebo (Gazebo as real world) | % improvement: 11.094 -> 18.93 | Helped |
| GAT vs no grounding (ablation) | CMA-ES in SimSpark · SimSpark to Gazebo (Gazebo as real world) | failures to find stable walk: 7 -> 1 | Helped |
| GAT vs Noise-Envelope on best generation index (ablation) | CMA-ES in SimSpark · SimSpark to Gazebo (Gazebo as real world) | avg generation of best policy: 6.6 -> 2.67 | Mixed |
| apply GAT with SimSpark (2 iterations) on real NAO | UNSW walk · Physical NAO walk | velocity (cm/s): 19.52 -> 27.97 | Helped |
| re-ground and optimize second iteration | GAT SimSpark · Physical NAO walk | velocity (cm/s): 26.27 -> 27.97 | Helped |
| use Gazebo instead of SimSpark for GAT | GAT · Physical NAO walk | velocity (cm/s): 26.27 -> 26.89 | No clear effect |
Closest papers by meaning
- Overcoming the Sim-to-Real Gap: Leveraging Simulation to Learn to Explore for Real-World RL (NeurIPS 2024 · similarity 0.85)
- Rapidly Adapting Policies to the Real-World via Simulation-Guided Fine-Tuning (ICLR 2025 · similarity 0.85)
- SUGAR: A Scalable Human-Video-Driven Generalizable Humanoid Loco-Manipulation Learning Framework (NeurIPS 2026 · similarity 0.85)
- Humanoid Locomotion as Next Token Prediction (NeurIPS 2024 · similarity 0.85)
- Humanoid Horizon: Extending Task Horizon in Whole-Body Loco-Manipulation via Parallel Training, Dynamic Starting, and Reward Gating (NeurIPS 2026 · similarity 0.85)
- Accelerating Visual-Policy Learning through Parallel Differentiable Simulation (NeurIPS 2025 · 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.