Counterfactual state explanations for reinforcement learning agents via generative deep learning

DOI10.1016/J.ARTINT.2021.103455OpenAlexW3124922852MaRDI QIDQ2238641FDOQ2238641

Authors: Matthew L. Olson, Roli Khanna, Lawrence Neal, Fuxin Li, Weng-Keen Wong 

Publication date: 2 November 2021

Published in: Artificial Intelligence (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2101.12446

Recommendations

• Local and global explanations of agent behavior: integrating strategy summaries with saliency maps
• Using state abstractions to compute personalized contrastive explanations for AI agent behavior
• A Symbolic Approach for Counterfactual Explanations
• A comparison of instance-level counterfactual explanation algorithms for behavioral and textual data: SEDC, LIME-C and SHAP-C
• An ASP-based approach to counterfactual explanations for classification

zbMATH Keywords

deep learningreinforcement learningexplainable AIinterpretable AI

Mathematics Subject Classification ID

Artificial intelligence (68Txx)

Cites Work

• Title not available (Why is that?)
• The book of why. The new science of cause and effect
• Counterfactuals.
• Interpolation coding: A representation for numbers in neural models
• Explanation in artificial intelligence: insights from the social sciences
• An Introduction to Variational Autoencoders

Cited In (3)

• Local and global explanations of agent behavior: integrating strategy summaries with saliency maps
• Using state abstractions to compute personalized contrastive explanations for AI agent behavior
• Counterexample explanation by learning small strategies in Markov decision processes

Uses Software

• OpenAI Gym
• PyTorch
• Adam
• Grad-CAM

This page was built for publication: Counterfactual state explanations for reinforcement learning agents via generative deep learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2238641)