Research map / Reinforcement learning / Visual and representation RL
LapGym - An Open Source Framework for Reinforcement Learning in Robot-Assisted Laparoscopic Surgery
Paul Maria Scheikl, Balázs Gyenes, Rayan Younis, Christoph Haas, Gerhard Neumann, Martin Wagner, Franziska Mathis-Ullrich
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 |
|---|---|---|---|
| Raise image resolution from 64x64 to 128x128 (ablation) | PPO image policy · TissueManipulationEnv (threshold 2 mm) | task success rate: 40% -> 100% | Improved |
| Raise image resolution from 64x64 to 128x128 (ablation) | PPO image policy · TissueManipulationEnv (threshold 5 mm) | learning speed: same final success, faster learning | Improved |
| Raise image resolution from 64x64 to 128x128 (ablation) | PPO image policy · ReachEnv | task success rate | No clear effect |
| Raise image resolution from 64x64 to 128x128 (ablation) | PPO image policy · DeflectSpheresEnv | task success rate: +roughly 5% | Mixed |
| Add a depth channel (RGBD instead of RGB) (ablation) | PPO image policy · All environments | task success rate: +5% to 10% | Improved |
| Add a depth channel (RGBD instead of RGB) (ablation) | PPO image policy · ReachEnv (radius 20 mm) | task success rate | Improved |
| Add a depth channel (RGBD instead of RGB) (ablation) | PPO image policy · ReachEnv (radius 8 mm) | task success rate | Mixed |
| Randomize path function parameters (ablation) | PPO · PrecisionCuttingEnv | task success rate: 95% -> 50% (image), 95% -> 60% (state) | Hurt |
| Control two instruments instead of one (ablation) | PPO state · DeflectSpheresEnv | task success rate | Hurt |
| Lower task threshold from 5 mm to 2 mm (ablation) | PPO image policy · TissueManipulationEnv | task success rate: 95% -> 40% | Hurt |
| Randomize landmark position on tissue (ablation) | PPO image policy · TissueManipulationEnv | learning speed | Hurt |
| Increase number of phases to learn (ablation) | PPO state · PickAndPlaceEnv and GraspLiftAndTouchEnv | task success rate: roughly 40% and 8% | Hurt |
Closest papers by meaning
- SurgicAI: A Hierarchical Platform for Fine-Grained Surgical Policy Learning and Benchmarking (NeurIPS 2024 · similarity 0.88)
- Deep Reinforcement Learning for Robotics: A Survey of Real-World Successes (AAAI 2025 · similarity 0.86)
- Efficient Language-instructed Skill Acquisition via Reward-Policy Co-Evolution (AAAI 2025 · similarity 0.85)
- Multi-Stage Manipulation with Demonstration-Augmented Reward, Policy, and World Model Learning (ICML 2025 · similarity 0.84)
- Plan-Seq-Learn: Language Model Guided RL for Solving Long Horizon Robotics Tasks (ICLR 2024 · similarity 0.84)
- LaST-R1: Reinforcing Robotic Manipulation via Adaptive Physical Latent Reasoning (NeurIPS 2026 · similarity 0.84)
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.
