Asymptotically Best Causal Effect Identification with Multi-Armed Bandits

Alan Malek, Silvia Chiappa

Keywords: causal identification formulas, frontdoor criterion, adjustment criterion, causal inference, multi armed bandits, best-arm-identification, confidence sequences, double machine learning, causal effect estimator, nuisance parameter

TL;DR: we propose using best-arm-identification bandits to select the best estimator for identifying a causal effect

Abstract: This paper considers the problem of selecting a formula for identifying a causal quantity of interest among a set of available formulas. We assume an online setting in which the investigator may alter the data collection mechanism in a data-dependent way with the aim of identifying the formula with lowest asymptotic variance in as few samples as possible. We formalize this setting by using the best-arm-identification bandit framework where the standard goal of learning the arm with the lowest loss is replaced with the goal of learning the arm that will produce the best estimate. We introduce new tools for constructing finite-sample confidence bounds on estimates of the asymptotic variance that account for the estimation of potentially complex nuisance functions, and adapt the best-arm-identification algorithms of LUCB and Successive Elimination to use these bounds. We validate our method by providing upper bounds on the sample complexity and an empirical study on artificially generated data.

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/deepmind/abcei_mab