Foundations of probability-raising causality in Markov decision processes

Research output: Contribution to journalResearch articleContributedpeer-review

Abstract

This work introduces a novel cause-effect relation in Markov decision processes using the probability-raising principle. Initially, sets of states as causes and effects are considered, which is subsequently extended to regular path properties as effects and then as causes. The paper lays the mathematical foundations and analyzes the algorithmic properties of these cause-effect relations. This includes algorithms for checking cause conditions given an effect and deciding the existence of probability-raising causes. As the definition allows for sub-optimal coverage properties, quality measures for causes inspired by concepts of statistical analysis are studied. These include recall, coverage ratio and f-score. The computational complexity for finding optimal causes with respect to these measures is analyzed.

Details

Original languageEnglish
Pages (from-to)4:1–4:66
Number of pages66
JournalLogical methods in computer science : LMCS
Volume20
Issue number1
Publication statusPublished - 19 Jan 2024
Peer-reviewedYes

External IDs

ORCID /0000-0002-5321-9343/work/155290604
ORCID /0000-0002-8490-1433/work/155291878
Scopus 85186474113
ORCID /0000-0003-4829-0476/work/165453939

Keywords

Keywords

  • probability-raising, Markov decision process, probabilistic causality, binary classifier

Library keywords