Research map / Reinforcement learning / Policy gradient and actor-critic
Safe Exploration for Efficient Policy Evaluation and Comparison
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 SEPEC instead of the Mixture policy | Bandit policy comparison · MAB, CMAB, linear bandits, and MNIST experiments | RMSE | Helped |
| Use SEPEC instead of the Safe Optimal Design policy | Bandit policy comparison · MAB, CMAB, linear bandits, and MNIST experiments | power | Helped |
| Use SEPEC instead of the Mixture or Safe Optimal Design policy | Bandit policy comparison · MAB, CMAB, and linear bandits | safety-violation loss | Helped |
| Use A/B testing or uniform exploration instead of SEPEC | Safe bandit policy comparison · Experiments at very low risk tolerance | RMSE | Mixed |
| Remove the safety constraint from SEPEC optimization (ablation) | SEPEC · MAB, CMAB, and linear bandits | RMSE | Helped |
| Relax the safety constraint by increasing risk tolerance | SEPEC · MAB, CMAB, and linear bandits | RMSE | Helped |
| Apply safety constraints in a linear bandit with a generalization function | SEPEC · Linear bandits | RMSE | Mixed |
Closest papers by meaning
- Doubly Optimal Policy Evaluation for Reinforcement Learning (ICLR 2025 · similarity 0.86)
- Efficient Policy Evaluation with Safety Constraint for Reinforcement Learning (ICLR 2025 · similarity 0.85)
- Adaptive Exploration for Multi-Reward Multi-Policy Evaluation (ICML 2025 · similarity 0.85)
- Infrequent Exploration in Linear Bandits (NeurIPS 2025 · similarity 0.85)
- Enhancing Efficiency of Safe Reinforcement Learning via Sample Manipulation (NeurIPS 2024 · similarity 0.85)
- SaVeR: Optimal Data Collection Strategy for Safe Policy Evaluation in Tabular MDP (ICML 2024 · 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.