Directional Domain Generalization

Abstract

Domain generalization aims to learn a predictive model from multiple different but related source tasks that can generalize well to a target task without the need of accessing any target data. Existing domain generalization methods ignore the relation between tasks, implicitly assuming that all the tasks are sampled from a stationary environment. Therefore, they can fail when deployed in an evolving environment. To this end, we formulate and study the directional domain generalization (DDG) scenario, which exploits not only the source data but also their evolving pattern to generate a model for the unseen task. Our theoretical result reveals the benefits of modeling the relation between two consecutive tasks by learning a globally consistent directional mapping function. In practice, our analysis also suggest solving the DDG problem in a meta-learning manner, which leads to directional prototypical network, the first method for the DDG problem. Empirical evaluation on both synthetic and real-world data sets validates the effectiveness of our approach.

Tasks

Reproductions