A causal framework for distribution generalization
Authors:
Preprint
Abstract
We consider the problem of predicting a response $Y$ from a set of covariates $X$ when test and training distributions differ. Since such differences may have causal explanations, we consider test distributions that emerge from interventions in a structural causal model, and focus on minimizing the worst-case risk. Causal regression models, which regress the response on its direct causes, remain unchanged under arbitrary interventions on the covariates, but they are not always optimal in the above sense. For example, for linear models and bounded interventions, alternative solutions have been shown to be minimax prediction optimal. We introduce the formal framework of distribution generalization that allows us to analyze the above problem in partially observed nonlinear models for both direct interventions on $X$ and interventions that occur indirectly via exogenous variables $A$. It takes into account that, in practice, minimax solutions need to be identified from data. Our framework allows us to characterize under which class of interventions the causal function is minimax optimal. We prove sufficient conditions for distribution generalization and present corresponding impossibility results. We propose a practical method, NILE, that achieves distribution generalization in a nonlinear IV setting with linear extrapolation. We prove consistency and present empirical results.
Keywords
All Related Versions
This publication has 21 references indexed in Scilit:
- Distributional robustness of K-class estimators and the PULSEThe Econometrics Journal, 2021
- Anchor Regression: Heterogeneous Data Meet CausalityJournal of the Royal Statistical Society Series B: Statistical Methodology, 2021
- Conditional variance penalties and domain shift robustnessMachine Learning, 2020
- Data-driven distributionally robust optimization using the Wasserstein metric: performance guarantees and tractable reformulationsMathematical Programming, 2017
- A Survey on Transfer LearningIEEE Transactions on Knowledge and Data Engineering, 2009
- Estimation of causal effects using linear non-Gaussian causal models with hidden variablesInternational Journal of Approximate Reasoning, 2008
- Instrumental Variable Estimation of Nonparametric ModelsEconometrica, 2003
- Finite-time Analysis of the Multiarmed Bandit ProblemMachine Learning, 2002
- Improving predictive inference under covariate shift by weighting the log-likelihood functionJournal of Statistical Planning and Inference, 2000
- Asymptotically efficient adaptive allocation rulesAdvances in Applied Mathematics, 1985