Causal Bayes Nets in Causal Learning and Inference

DOI Web Site Open Access
  • Saito Motoyuki
    Department of Psychology, University of California, Los Angeles

Bibliographic Information

Other Title
  • 因果性の学習と推論における因果ベイズネットについて
  • インガセイ ノ ガクシュウ ト スイロン ニ オケル インガ ベイズネット ニ ツイテ

Search this article

Abstract

 Causal knowledge enables us to explain past events, to control present environment,<br>and to predict future outcomes. Over the last decade, causal Bayes nets have been rec-<br>ognized as a normative framework for causality and used as a psychological model to<br> account for human causal learning and inference. This article provides an introduction<br> to causal Bayes nets. According to causal Bayes nets, causal inference can be divided<br> into three processes: (a) learning the structure of the causal network, (b) learning the<br> strength of the causal relations, and (c) inferring the effect from the cause or the cause<br> from the effect. For each process, I describe the predictions of causal Bayes nets, review<br> experimental results, and suggest future directions. Although there are a few excep-<br>tions (e.g., Markov violation), most of the results are consistent with the predictions<br> of causal Bayes nets. The current problems of the Bayesian approach and its future<br> perspective are discussed.

Journal

Related Projects

See more

Keywords

Details 詳細情報について

Report a problem

Back to top