Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Planning and Acting with Non-Deterministic Events: Navigating between Safe States
Lukas Chrpa, Jakub Gemrot, Martin Pilat
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 |
|---|---|---|---|
| limit consecutive steps in unsafe states (LIMIT) vs dangerousness-based DANG | PER replanning · Perestroika-5 | successful runs: 83 -> 100 | Improved |
| LIMIT vs DANG | PER replanning · AUV-3 | successful runs: 80 -> 100 | Improved |
| LIMIT vs PER-APP | PER replanning · Perestroika-5 | successful runs: 3 -> 100 | Improved |
| LIMIT vs PER-EVENT | PER replanning · Perestroika-5 | successful runs: 1 -> 100 | Improved |
| LIMIT vs DANG | PER replanning · Perestroika-5 | planning time (ms): 21479 -> 2198 | Improved |
| LIMIT vs DANG | PER replanning · Perestroika-6 | planning time (ms): 14170 -> 633 | Improved |
| LIMIT vs DANG | PER replanning · AUV-3 | planning time (ms): 1620 -> 2198 | Hurt |
| LIMIT vs DANG | PER replanning · AUV-6 | planning time (ms): 2681 -> 6982 | Hurt |
| LIMIT vs DANG | PER replanning · Perestroika-6 | actions: 111.78 -> 15.84 | Improved |
| LIMIT vs DANG | PER replanning · AUV-1 | actions: 41.81 -> 38.60 | No clear effect |
| encode as FOND and use PRP strong cyclic planner instead of LIMIT | PRP · AUV-1 | solved problems: 987 seconds | Hurt |
Closest papers by meaning
- Building a Foundational Guardrail for General Agentic Systems via Synthetic Data (ICLR 2026 · similarity 0.85)
- Learning Planning Domains from Non-redundant Fully-Observed Traces: Theoretical Foundations and Complexity Analysis (AAAI 2024 · similarity 0.85)
- Safe Planner: Empowering Safety Awareness in Large Pre-Trained Models for Robot Task Planning (AAAI 2025 · similarity 0.84)
- PRP Rebooted: Advancing the State of the Art in FOND Planning (AAAI 2024 · similarity 0.84)
- Predicting Future Actions of Reinforcement Learning Agents (NeurIPS 2024 · similarity 0.84)
- Safeguarded Progress in Reinforcement Learning: Safe Bayesian Exploration for Control Policy Synthesis (AAAI 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.
