Justifying information-geometric causal inference

DOI10.1007/978-3-319-21852-6_18zbMATH Open1337.62190arXiv1402.2499OpenAlexW2304588021MaRDI QIDQ2805731FDOQ2805731

Authors: Bastian Steudel, Naji Shajarisales, Bernhard Schölkopf, Dominik Janzing 

Publication date: 13 May 2016

Published in: Measures of Complexity (Search for Journal in Brave)

Abstract: Information Geometric Causal Inference (IGCI) is a new approach to distinguish between cause and effect for two variables. It is based on an independence assumption between input distribution and causal mechanism that can be phrased in terms of orthogonality in information space. We describe two intuitive reinterpretations of this approach that makes IGCI more accessible to a broader audience. Moreover, we show that the described independence is related to the hypothesis that unsupervised learning and semi-supervised learning only works for predicting the cause from the effect and not vice versa.

Full work available at URL: https://arxiv.org/abs/1402.2499

Recommendations

• Information-geometric approach to inferring causal directions
• Justifying additive noise model-based causal discovery via algorithmic information theory
• From dependency to causality: a machine learning approach
• Distinguishing cause from effect using observational data: methods and benchmarks
• Semi-supervised interpolation in an anticausal learning scenario

Mathematics Subject Classification ID

Statistical aspects of information-theoretic topics (62B10) Linear inference, regression (62J99)

Cites Work

• Causation, prediction, and search
• Title not available (Why is that?)
• Estimation of Dependences Based on Empirical Data
• Information-geometric approach to inferring causal directions
• Causal Inference Using the Algorithmic Markov Condition

Cited In (3)

• Semi-supervised interpolation in an anticausal learning scenario
• Information-geometric approach to inferring causal directions
• Unified framework for information integration based on information geometry

Uses Software

• TETRAD

This page was built for publication: Justifying information-geometric causal inference

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2805731)