Learning Generalized Gumbel-max Causal Mechanisms

Guy Lorberbom, Daniel D. Johnson, Chris J. Maddison, Daniel Tarlow, Tamir Hazan

Keywords: Gumbel max, reparameterization trick, structural causal model

TL;DR: Learn a causal mechanism that minimizes a quantitative criteria (such as variance) when used in a counterfactual context.

Abstract: To perform counterfactual reasoning in Structural Causal Models (SCMs), one needs to know the causal mechanisms, which provide factorizations of conditional distributions into noise sources and deterministic functions mapping realizations of noise to samples. Unfortunately, the causal mechanism is not uniquely identified by data that can be gathered by observing and interacting with the world, so there remains the question of how to choose causal mechanisms. In recent work, Oberst & Sontag (2019) propose Gumbel-max SCMs, which use Gumbel-max reparameterizations as the causal mechanism due to an appealing counterfactual stability property. However, the justification requires appealing to intuition. In this work, we instead argue for choosing a causal mechanism that is best under a quantitative criteria such as minimizing variance when estimating counterfactual treatment effects. We propose a parameterized family of causal mechanisms that generalize Gumbel-max. We show that they can be trained to minimize counterfactual effect variance and other losses on a distribution of queries of interest, yielding lower variance estimates of counterfactual treatment effect than fixed alternatives, also generalizing to queries not seen at training time.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

Code: https://github.com/google-research/google-research/tree/master/gumbel_max_causal_gadgets

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/learning-generalized-gumbel-max-causal/code)