Predictors from causal features do not generalize better to new domains

arXiv:2402.09891v1 Announce Type: new
Abstract: We study how well machine learning models trained on causal features generalize across domains. We consider 16 prediction tasks on tabular datasets covering applications in health, employment, education, social benefits, and politics. Each dataset comes with multiple domains, allowing us to test how well a model trained in one domain performs in another. For each prediction task, we select features that have a causal influence on the target of prediction. Our goal is to test …

abstract applications arxiv benefits cs.lg dataset datasets domains education employment features health machine machine learning machine learning models multiple politics prediction social stat.ml study tabular tasks test type