Research map / Reinforcement learning / Policy gradient and actor-critic
Safe Policy Improvement with Baseline Bootstrapping in Factored Environments
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 factored representation in Pi_b-SPIBB | Pi_b-SPIBB · Taxi, softmax baseline, |D|=50 | mean return: -24.36 -> -5.94 | Helped |
| use factored representation in Pi_<=b-SPIBB | Pi_<=b-SPIBB · Taxi, softmax baseline, |D|=50 | mean return: -16.6 -> -1.08 | Helped |
| use factored representation in Pi_b-SPIBB | Pi_b-SPIBB · Taxi, softmax baseline, |D|=50 | 1%-quantile return: -26.54 -> -9.72 | Helped |
| use factored representation in Pi_<=b-SPIBB | Pi_<=b-SPIBB · Taxi, softmax baseline, |D|=50 | 1%-quantile return: -19.7 -> -4.32 | Helped |
| add safety bootstrapping (SPIBB) to basic factored RL | Factored RL · Taxi, softmax baseline, |D|=50 | mean return: 5.56 -> -1.08 | Hurt |
| relax Pi_b to Pi_<=b constraint (flat) (ablation) | Pi_b-SPIBB · Taxi, softmax baseline, |D|=50 | mean return: -24.36 -> -16.6 | Helped |
| factored representation with deterministic behavior policy | Pi_b-SPIBB · SysAdmin, deterministic baseline | policy improvement over baseline | Helped |
Closest papers by meaning
- Enhancing Efficiency of Safe Reinforcement Learning via Sample Manipulation (NeurIPS 2024 · similarity 0.88)
- Scalable Safe Policy Improvement for Factored Multi-Agent MDPs (ICML 2024 · similarity 0.88)
- Long-Term Safe Reinforcement Learning with Binary Feedback (AAAI 2024 · similarity 0.87)
- Safety Representations for Safer Policy Learning (ICLR 2025 · similarity 0.87)
- Sampling-Based Safe Reinforcement Learning (NeurIPS 2026 · similarity 0.87)
- Offline Safe Reinforcement Learning Using Trajectory Classification (AAAI 2025 · similarity 0.87)
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.