Single World Intervention Graphs (SWIGs): A Unification of the Counterfactual and Graphical Approaches to Causality - Foundations and Trends in Machine Learning (Paperback)
  • Single World Intervention Graphs (SWIGs): A Unification of the Counterfactual and Graphical Approaches to Causality - Foundations and Trends in Machine Learning (Paperback)
zoom

Single World Intervention Graphs (SWIGs): A Unification of the Counterfactual and Graphical Approaches to Causality - Foundations and Trends in Machine Learning (Paperback)

(author), (author)
£92.00
Paperback 195 Pages / Published: 28/06/2017
  • We can order this

Usually dispatched within 2 weeks

  • This item has been added to your basket
This book presents a simple graphical theory unifying causal directed acyclic graphs (DAGs) and potential (aka counterfactual) outcomes via a node-splitting transformation. It introduces a new graph, the Single-World Intervention Graph (SWIG). The SWIG encodes the counterfactual independences associated with a specific hypothetical intervention on the set of treatment variables. The nodes on the SWIG are the corresponding counterfactual random variables. The theory is illustrated with a number of examples. The authors' graphical theory of SWIGs may be used to infer the counterfactual independence relations implied by the counterfactual models developed in Robins (1986, 1987). Moreover, in the absence of hidden variables, the joint distribution of the counterfactuals is identified; the identifying formula is the extended g-computation formula introduced in (Robins et al., 2004). Although Robins (1986, 1987) did not use DAGs we translate his algebraic results to facilitate understanding of this prior work. An attractive feature of Robins' approach that it largely avoids making counterfactual independence assumptions that are experimentally untestable.As an important illustration, this book revisits the critique of Robins' g-computation given in (Pearl, 2009, Ch. 11.3.7); it uses SWIGs to show that all of Pearl's claims are either erroneous or based on misconceptions. It also shows that simple extensions of the formalism may be used to accommodate dynamic regimes, and to formulate non-parametric structural equation models in which assumptions relating to the absence of direct effects are formulated at the population level. Finally, this book shows that the authors' graphical theory also naturally arises in the context of an expanded causal Bayesian network in which we are able to observe the natural state of a variable prior to intervention.

Publisher: now publishers Inc
ISBN: 9781601988102
Number of pages: 195

You may also be interested in...

Machine Learning
Added to basket
£52.99
Mixed media product
Probabilistic Graphical Models
Added to basket
Bayesian Reasoning and Machine Learning
Added to basket
Machine Learning for Hackers
Added to basket
Machine Learning
Added to basket
Machine Learning
Added to basket
£37.99
Paperback
Thoughtful Machine Learning
Added to basket
Reinforcement Learning
Added to basket
The Elements of Statistical Learning
Added to basket
Superintelligence
Added to basket
£18.99
Hardback
Web Data Mining
Added to basket
£54.99
Hardback
Machine Learning
Added to basket
£39.99
Paperback
Introduction to Machine Learning
Added to basket

Reviews

Please sign in to write a review

Your review has been submitted successfully.