Action-model acquisition for planning via transfer learning

DOI10.1016/J.ARTINT.2014.03.004MaRDI QIDQ2453728zbMATH OpenOpenAlexWikidataFDO

Authors Hankz Hankui Zhuo, Qiang Yang

Publication date 10 June 2014

Published in Artificial Intelligence (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.artint.2014.03.004

zbMATH Keywords

action model learningAI planningdomain model acquisitationMarkov logic networkssimilarity measurestransfer learningweb search

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Problem solving in the context of artificial intelligence (heuristics, search strategies, etc.) (68T20)

Review text

AI planning represents an area of research in artificial intelligence with applications in a variety of domains, including robotics, process planning, web-based information gathering, etc. Many techniques involved in AI planning need action models to acquire sufficient training data to ensure the learned action models are appropriate. This paper deals with a novel algorithmic framework, namely TRAMP (TRansfer learning Action Models for Planning), which seeks to learn action models in the target domain by transferring knowledge from source domains, assuming they are already created and ready to be transferred. After the presentation of TRAMP, the authors empirically evaluate it and, finally, review previous related works and present their future work.

Recommendations

• Learning action models from plan examples using weighted MAX-SAT
• Action knowledge acquisition with opmaker2
• Learning action models with minimal observability
• Learning partially observable non-deterministic action models
• Learning action strategies for planning domains

Cites work

• scientific article; zbMATH DE number 5547890 (Why is no real title available?)
• scientific article; zbMATH DE number 2201583 (Why is no real title available?)
• Learning action models from plan examples using weighted MAX-SAT
• Learning recursive control programs from problem solving
• Markov logic networks
• On Information and Sufficiency
• STRIPS: A new approach to the application of theorem proving to problem solving
• The structure-mapping engine: Algorithm and examples
• Transfer learning for reinforcement learning domains: a survey

Cited in

(12)

• Learning action models from plan examples using weighted MAX-SAT
• Gradient-based mixed planning with symbolic and numeric action parameters
• A constraint-based approach to learn temporal features on action models from multiple plans
• Learning action models with minimal observability
• Model-lite planning: case-based vs. model-based approaches
• Learning hierarchical task network domains from partially observed plan traces
• scientific article; zbMATH DE number 5547939 (Why is no real title available?)
• Cost-sensitive action model learning
• Relational reinforcement learning for planning with exogenous effects
• Action knowledge acquisition with opmaker2
• Robust planning with incomplete domain models
• Using inductive rule learning techniques to learn planning domains

Describes a project that uses

Uses Software

• TRAMP

This page was built for publication: Action-model acquisition for planning via transfer learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2453728)