Research map / Reinforcement learning / MDPs and planning
Constrained and Robust Policy Synthesis with Satisfiability-Modulo-Probabilistic-Model-Checking
Linus Heck, Filip Macák, Milan Češka, Sebastian Junges
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 |
|---|---|---|---|
| use SMPMC (SMT solver with a probabilistic-model-checking theory) instead of a monolithic SMT(LRA) encoding | robust decision-tree policy synthesis (C1) · 112 benchmarks, 30-min timeout | benchmarks solved: 24 -> 77 | Improved |
| use SMPMC instead of SMT(LRA) | synthesis on instances SMT(LRA) can solve · C1 | runtime: 20x faster on average | Improved |
| use SMPMC instead of PAYNT-CEGIS | decision tree / constrained synthesis (C2) · 72 benchmarks | benchmarks solved: 54 vs 54 | No clear effect |
| use SMPMC instead of extended PAYNT-AR | robust synthesis (C3) · 82 benchmarks | benchmarks solved: 43 vs 43 | No clear effect |
| use SMPMC instead of PAYNT-AR | plain synthesis (C4) · 106 benchmarks | benchmarks solved: 88 vs 86 | No clear effect |
| use SMPMC instead of PAYNT-CEGIS/Hybrid | plain synthesis (C4) · 106 benchmarks | benchmarks solved: 88 vs 24 and 23 | Improved |
| restrict conflict clauses to parameters the reachable colorings depend on | SMPMC theory solver · Rocks, Refuel etc. | conflict size | Improved |
Closest papers by meaning
- Efficient Policy Optimization in Robust Constrained MDPs with Iteration Complexity Guarantees (NeurIPS 2025 · similarity 0.90)
- Provable Policy Gradient for Robust Average-Reward MDPs Beyond Rectangularity (ICML 2025 · similarity 0.89)
- Efficient Value Iteration for s-rectangular Robust Markov Decision Processes (ICML 2024 · similarity 0.88)
- Near-Optimal Policy Identification in Robust Constrained Markov Decision Processes via Epigraph Form (ICLR 2025 · similarity 0.88)
- Solving Robust Markov Decision Processes: Generic, Reliable, Efficient (AAAI 2025 · similarity 0.88)
- 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.
