Deep Backtracking Counterfactuals for Causally Compliant Explanations
This work addresses the need for causally compliant explanations in high-dimensional data, offering a modular alternative to existing counterfactual methods, though it appears incremental as it builds on less studied backtracking interpretations.
The authors tackled the problem of generating backtracking counterfactuals in structural causal models with deep generative components, introducing DeepBC with two versions (Langevin Monte Carlo and constrained optimization) that demonstrated causally compliant and versatile properties on modified MNIST and CelebA datasets.
Counterfactuals answer questions of what would have been observed under altered circumstances and can therefore offer valuable insights. Whereas the classical interventional interpretation of counterfactuals has been studied extensively, backtracking constitutes a less studied alternative where all causal laws are kept intact. In the present work, we introduce a practical method called deep backtracking counterfactuals (DeepBC) for computing backtracking counterfactuals in structural causal models that consist of deep generative components. We propose two distinct versions of our method--one utilizing Langevin Monte Carlo sampling and the other employing constrained optimization--to generate counterfactuals for high-dimensional data. As a special case, our formulation reduces to methods in the field of counterfactual explanations. Compared to these, our approach represents a causally compliant, versatile and modular alternative. We demonstrate these properties experimentally on a modified version of MNIST and CelebA.