Differentiable Causal Backdoor Discovery
This work addresses a key challenge in decision-making for domains like drug development and policy-making, but it appears incremental as it builds on existing methods for causal discovery.
The paper tackles the problem of discovering causal effects from observational data when confounders are present, by introducing a gradient-based optimization algorithm that uses auxiliary variables to find appropriate adjustments, and demonstrates that it outperforms practical alternatives in estimating the true causal effect.
Discovering the causal effect of a decision is critical to nearly all forms of decision-making. In particular, it is a key quantity in drug development, in crafting government policy, and when implementing a real-world machine learning system. Given only observational data, confounders often obscure the true causal effect. Luckily, in some cases, it is possible to recover the causal effect by using certain observed variables to adjust for the effects of confounders. However, without access to the true causal model, finding this adjustment requires brute-force search. In this work, we present an algorithm that exploits auxiliary variables, similar to instruments, in order to find an appropriate adjustment by a gradient-based optimization method. We demonstrate that it outperforms practical alternatives in estimating the true causal effect, without knowledge of the full causal graph.