Search and explore: symbiotic policy synthesis in POMDPs

DOI10.1007/978-3-031-37709-9_6MaRDI QIDQ6535683zbMATH OpenFDO

Authors Roman Andriushchenko, Milan Češka, Sebastian Junges, Joost-Pieter Katoen, Filip Macák

Publication date 1 February 2024

Copyright license Creative Commons Attribution 4.0 International

Mathematics Subject Classification ID

Specification and verification (program logics, model checking, etc.) (68Q60) Markov and semi-Markov decision processes (90C40)

Recommendations

• Enforcing almost-sure reachability in POMDPs
• Supervisor synthesis of POMDP via automata learning
• Constraint-based controller synthesis in non-deterministic and partially observable domains
• Task-aware verifiable RNN-based policies for partially observable Markov decision processes
• Planning and acting in partially observable stochastic domains

Cites work

• scientific article; zbMATH DE number 700091 (Why is no real title available?)
• A practitioner's guide to MDP model checking algorithms
• Gradient-descent for randomized controllers under partial observability
• Inductive synthesis for probabilistic programs reaches new horizons
• On the undecidability of probabilistic planning and related stochastic optimization problems
• Shepherding hordes of Markov chains
• Task-aware verifiable RNN-based policies for partially observable Markov decision processes
• The Optimal Control of Partially Observable Markov Processes over a Finite Horizon
• Under-approximating expected total rewards in POMDPs
• Verification and control of partially observable probabilistic systems
• Verification of Indefinite-Horizon POMDPs

This page was built for publication: Search and explore: symbiotic policy synthesis in POMDPs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6535683)