Approximate policy iteration with a policy language bias: solving relational Markov decision processes

MaRDI QIDQ3623997zbMATH OpenFDO

Authors Alan Fern, Sungwook Yoon, Robert L. Givan

Publication date 28 April 2009

Full work available at URL https://arxiv.org/abs/1109.2156

Mathematics Subject Classification ID

Problem solving in the context of artificial intelligence (heuristics, search strategies, etc.) (68T20)

Recommendations

• Practical solution techniques for first-order MDPs
• Finding memoryless probabilistic relational policies for inter-task reuse
• First order decision diagrams for relational MDPS
• Inductive Logic Programming
• Relational reinforcement learning

Cited in

(14)

• Structured machine learning: the next ten years
• Preference-based reinforcement learning: a formal framework and a policy iteration algorithm
• Rollout sampling approximate policy iteration
• Finding memoryless probabilistic relational policies for inter-task reuse
• Bounds for Multistage Stochastic Programs Using Supervised Learning Strategies
• A Comprehensive Framework for Learning Declarative Action Models
• Qualitative numeric planning: reductions and complexity
• A new representation and associated algorithms for generalized planning
• Reducing reinforcement learning to KWIK online regression
• Generalized planning as heuristic search: a new planning search-space that leverages pointers over objects
• Deep reinforcement learning for approximate policy iteration: convergence analysis and a post-earthquake disaster response case study
• The factored policy-gradient planner
• Practical solution techniques for first-order MDPs
• APPSSAT: Approximate probabilistic planning using stochastic satisfiability

This page was built for publication: Approximate policy iteration with a policy language bias: solving relational Markov decision processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3623997)