Zero-shot causal learning

TL;DR

This paper introduces CaML, a meta-learning framework for zero-shot causal prediction, enabling personalized effect estimation of novel interventions without prior data, with applications in medicine and biology.

Contribution

The paper presents a novel causal meta-learning approach that predicts effects of unseen interventions by training on diverse tasks, outperforming traditional methods.

Findings

CaML effectively predicts effects of new interventions in real-world datasets.

Zero-shot predictions outperform baselines trained on intervention-specific data.

The approach leverages intervention attributes and individual features for personalized predictions.

Abstract

Predicting how different interventions will causally affect a specific individual is important in a variety of domains such as personalized medicine, public policy, and online marketing. There are a large number of methods to predict the effect of an existing intervention based on historical data from individuals who received it. However, in many settings it is important to predict the effects of novel interventions (e.g., a newly invented drug), which these methods do not address. Here, we consider zero-shot causal learning: predicting the personalized effects of a novel intervention. We propose CaML, a causal meta-learning framework which formulates the personalized prediction of each intervention's effect as a task. CaML trains a single meta-model across thousands of tasks, each constructed by sampling an intervention, its recipients, and its nonrecipients. By leveraging both intervention information (e.g., a drug's attributes) and individual features~(e.g., a patient's history), CaML is able to predict the personalized effects of novel interventions that do not exist at the time of training. Experimental results on real world datasets in large-scale medical claims and cell-line perturbations demonstrate the effectiveness of our approach. Most strikingly, \method's zero-shot predictions outperform even strong baselines trained directly on data from the test interventions.