A human-centered data-driven planner-actor-critic architecture via logic programming

MaRDI QIDQ5020557zbMATH OpenFDO

Authors Daoming Lyu, Fangkai Yang, Bo Liu, Steven Gustafson

Publication date 6 January 2022

Full work available at URL https://arxiv.org/abs/1909.09209, http://eptcs.web.cse.unsw.edu.au/paper.cgi?ICLP2019.23

Mathematics Subject Classification ID

Logic programming (68N17)

Recommendations

• Reinforcement learning
• Learning reward machines: a study in partially observable reinforcement learning
• Classical Planning in Deep Latent Space
• Specification-guided reinforcement learning
• scientific article; zbMATH DE number 2033218

Cites work

• A synthesis of automated planning and reinforcement learning for efficient, robust decision-making
• Between MDPs and semi-MDPs: A framework for temporal abstraction in reinforcement learning
• Generality in artificial intelligence
• Mobile robot planning using action language \({\mathcal {BC}}\) with an abstraction hierarchy
• Natural actor-critic algorithms
• Planning and acting in partially observable stochastic domains
• Recent advances in hierarchical reinforcement learning
• Reinforcement learning. An introduction
• Robot task planning and explanation in open and uncertain worlds
• Simple statistical gradient-following algorithms for connectionist reinforcement learning
• Some properties of system descriptions of \(\mathcal{AL}_d\)
• The fast downward planning system

Cited in

(1)

• Verifiably safe exploration for end-to-end reinforcement learning

Describes a project that uses

Uses Software

• ALM
• PEORL
• BWIBots

This page was built for publication: A human-centered data-driven planner-actor-critic architecture via logic programming

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5020557)