Class-Incremental Domain Adaptation

TL;DR

This paper introduces Class-Incremental Domain Adaptation (CIDA), a new paradigm that combines domain adaptation and class-incremental learning to classify both shared and novel classes under domain shift.

Contribution

It proposes a prototypical network-inspired method for CIDA, addressing limitations of existing DA and CI approaches under domain shift.

Findings

Outperforms existing DA and CI methods in CIDA tasks.

Effectively classifies shared and novel classes under domain shift.

Demonstrates superior performance in empirical evaluations.

Abstract

We introduce a practical Domain Adaptation (DA) paradigm called Class-Incremental Domain Adaptation (CIDA). Existing DA methods tackle domain-shift but are unsuitable for learning novel target-domain classes. Meanwhile, class-incremental (CI) methods enable learning of new classes in absence of source training data but fail under a domain-shift without labeled supervision. In this work, we effectively identify the limitations of these approaches in the CIDA paradigm. Motivated by theoretical and empirical observations, we propose an effective method, inspired by prototypical networks, that enables classification of target samples into both shared and novel (one-shot) target classes, even under a domain-shift. Our approach yields superior performance as compared to both DA and CI methods in the CIDA paradigm.