Skip to yearly menu bar Skip to main content


In-Person Poster presentation / poster accept

Measuring axiomatic soundness of counterfactual image models

Miguel Monteiro · Fabio De Sousa Ribeiro · Nick Pawlowski · Daniel Castro · Ben Glocker

MH1-2-3-4 #101

Keywords: [ generative models ] [ computer vision ] [ Counterfactual inference ] [ Probabilistic Methods ]


Abstract:

We present a general framework for evaluating image counterfactuals. The power and flexibility of deep generative models make them valuable tools for learning mechanisms in structural causal models. However, their flexibility makes counterfactual identifiability impossible in the general case.Motivated by these issues, we revisit Pearl's axiomatic definition of counterfactuals to determine the necessary constraints of any counterfactual inference model: composition, reversibility, and effectiveness. We frame counterfactuals as functions of an input variable, its parents, and counterfactual parents and use the axiomatic constraints to restrict the set of functions that could represent the counterfactual, thus deriving distance metrics between the approximate and ideal functions. We demonstrate how these metrics can be used to compare and choose between different approximate counterfactual inference models and to provide insight into a model's shortcomings and trade-offs.

Chat is not available.