Research map / Reinforcement learning / Policy gradient and actor-critic
Conservative Safety Critics for Exploration
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 |
|---|---|---|---|
| Use a conservative safety critic and reject actions whose estimated failure probability exceeds a threshold | CSC · Five simulated robotic control domains | catastrophic failures during training: up to 50% | Helped |
| Use CSC instead of prior safe exploration methods | CSC · Five simulated robotic control domains | task performance | Mixed |
| Use a stricter safety threshold (ablation) | CSC · Five simulated robotic control domains | average failures per episode | Helped |
| Use a stricter safety threshold (ablation) | CSC · Five simulated robotic control domains | task reward convergence rate | Hurt |
Closest papers by meaning
- Safeguarded Progress in Reinforcement Learning: Safe Bayesian Exploration for Control Policy Synthesis (AAAI 2024 · similarity 0.92)
- Sampling-Based Safe Reinforcement Learning (NeurIPS 2026 · similarity 0.91)
- Safety Representations for Safer Policy Learning (ICLR 2025 · similarity 0.91)
- Balance Reward and Safety Optimization for Safe Reinforcement Learning: A Perspective of Gradient Manipulation (AAAI 2024 · similarity 0.90)
- Dyna-Style Safety Augmented Reinforcement Learning: Staying Safe in the Face of Uncertainty (NeurIPS 2026 · similarity 0.90)
- Efficient Policy Evaluation with Safety Constraint for Reinforcement Learning (ICLR 2025 · similarity 0.89)
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.