Learning Disentangled Semantic Representation for Domain Adaptation

TL;DR

This paper introduces a novel approach for domain adaptation by disentangling semantic and domain information in latent space using a variational auto-encoder and adversarial training, achieving state-of-the-art results.

Contribution

It proposes a disentangled semantic representation method that separates domain-invariant semantics from domain-specific features for improved adaptation.

Findings

Achieves state-of-the-art performance on domain adaptation benchmarks.

Effectively disentangles semantic and domain features in latent space.

Improves cross-domain semantic transfer accuracy.

Abstract

Domain adaptation is an important but challenging task. Most of the existing domain adaptation methods struggle to extract the domain-invariant representation on the feature space with entangling domain information and semantic information. Different from previous efforts on the entangled feature space, we aim to extract the domain invariant semantic information in the latent disentangled semantic representation (DSR) of the data. In DSR, we assume the data generation process is controlled by two independent sets of variables, i.e., the semantic latent variables and the domain latent variables. Under the above assumption, we employ a variational auto-encoder to reconstruct the semantic latent variables and domain latent variables behind the data. We further devise a dual adversarial network to disentangle these two sets of reconstructed latent variables. The disentangled semantic latent variables are finally adapted across the domains. Experimental studies testify that our model yields state-of-the-art performance on several domain adaptation benchmark datasets.