Ensembles of Locally Independent Prediction Models

TL;DR

This paper introduces a new method for training ensemble models that focus on local prediction diversity, improving generalization especially under covariate shift and data limitations.

Contribution

It proposes a novel diversity metric and training approach that emphasizes local extrapolation differences, addressing limitations of existing ensemble methods.

Findings

Improved generalization under covariate shift.

Enhanced diversity in ensemble predictions.

Effective on both synthetic and real-world tasks.

Abstract

Ensembles depend on diversity for improved performance. Many ensemble training methods, therefore, attempt to optimize for diversity, which they almost always define in terms of differences in training set predictions. In this paper, however, we demonstrate the diversity of predictions on the training set does not necessarily imply diversity under mild covariate shift, which can harm generalization in practical settings. To address this issue, we introduce a new diversity metric and associated method of training ensembles of models that extrapolate differently on local patches of the data manifold. Across a variety of synthetic and real-world tasks, we find that our method improves generalization and diversity in qualitatively novel ways, especially under data limits and covariate shift.