Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation

TL;DR

This paper introduces DRCN, a deep learning model that jointly learns to classify labeled source data and reconstruct unlabeled target data, improving cross-domain object recognition accuracy.

Contribution

The paper presents a novel unsupervised domain adaptation model, DRCN, which combines classification and reconstruction to enhance feature transfer across domains.

Findings

DRCN improves accuracy by up to ~8% over previous methods.

Reconstruction transforms source images to resemble target domain images.

The shared representation encodes both target structure and source labels.

Abstract

In this paper, we propose a novel unsupervised domain adaptation algorithm based on deep learning for visual object recognition. Specifically, we design a new model called Deep Reconstruction-Classification Network (DRCN), which jointly learns a shared encoding representation for two tasks: i) supervised classification of labeled source data, and ii) unsupervised reconstruction of unlabeled target data.In this way, the learnt representation not only preserves discriminability, but also encodes useful information from the target domain. Our new DRCN model can be optimized by using backpropagation similarly as the standard neural networks. We evaluate the performance of DRCN on a series of cross-domain object recognition tasks, where DRCN provides a considerable improvement (up to ~8% in accuracy) over the prior state-of-the-art algorithms. Interestingly, we also observe that the reconstruction pipeline of DRCN transforms images from the source domain into images whose appearance resembles the target dataset. This suggests that DRCN's performance is due to constructing a single composite representation that encodes information about both the structure of target images and the classification of source images. Finally, we provide a formal analysis to justify the algorithm's objective in domain adaptation context.