From Big to Small: Adaptive Learning to Partial-Set Domains

TL;DR

This paper introduces Partial Domain Adaptation, a new paradigm that relaxes class space assumptions, and proposes a selective adversarial network with a bi-level strategy to improve adaptation from large to small domains.

Contribution

It provides a theoretical analysis of partial domain adaptation and develops SAN++, a novel model with bi-level selection for better transfer learning.

Findings

SAN++ outperforms existing methods on standard datasets.

Theoretical analysis highlights importance of estimating transferable probabilities.

Bi-level selection improves class and instance transferability.

Abstract

Domain adaptation targets at knowledge acquisition and dissemination from a labeled source domain to an unlabeled target domain under distribution shift. Still, the common requirement of identical class space shared across domains hinders applications of domain adaptation to partial-set domains. Recent advances show that deep pre-trained models of large scale endow rich knowledge to tackle diverse downstream tasks of small scale. Thus, there is a strong incentive to adapt models from large-scale domains to small-scale domains. This paper introduces Partial Domain Adaptation (PDA), a learning paradigm that relaxes the identical class space assumption to that the source class space subsumes the target class space. First, we present a theoretical analysis of partial domain adaptation, which uncovers the importance of estimating the transferable probability of each class and each instance across domains. Then, we propose Selective Adversarial Network (SAN and SAN++) with a bi-level selection strategy and an adversarial adaptation mechanism. The bi-level selection strategy up-weighs each class and each instance simultaneously for source supervised training, target self-training, and source-target adversarial adaptation through the transferable probability estimated alternately by the model. Experiments on standard partial-set datasets and more challenging tasks with superclasses show that SAN++ outperforms several domain adaptation methods.