Research map / Reinforcement learning / MDPs and planning
Safe Exploration and Optimization of Constrained MDPs Using Gaussian Processes
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 |
|---|---|---|---|
| interpolate optimistic and pessimistic safe MDP policies (SafeExpOpt-MDP) vs Moldovan & Abbeel | safe exploration · random grid world | normalized cumulative reward: 0.740 ± 0.213 -> 0.774 ± 0.183 | Helped |
| SafeExpOpt-MDP vs Turchetta et al. safe exploration | safe exploration · random grid world | percentage explored: 10.6 ± 2.1 -> 7.4 ± 2.9 | Helped |
| SafeExpOpt-MDP vs Moldovan & Abbeel | safe exploration · Mars scenario 1 | cumulative reward: 29 -> 36 | Helped |
| SafeExpOpt-MDP vs Moldovan & Abbeel | safe exploration · Mars scenario 2 | cumulative reward: 45 -> 67 | Helped |
Closest papers by meaning
- Safe Reinforcement Learning in Constrained Markov Decision Processes (ICML 2020 · similarity 0.94)
- Safe Exploration in Finite Markov Decision Processes with Gaussian Processes (NeurIPS 2016 · similarity 0.94)
- Constrained Markov Decision Processes via Backward Value Functions (ICML 2020 · similarity 0.90)
- Safeguarded Progress in Reinforcement Learning: Safe Bayesian Exploration for Control Policy Synthesis (AAAI 2024 · similarity 0.90)
- Probabilistic Shielding for Safe Reinforcement Learning (AAAI 2025 · similarity 0.89)
- Near-Optimal Sample Complexity for Online Constrained MDPs (NeurIPS 2025 · similarity 0.88)
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.