Task Guided Compositional Representation Learning for ZDA

TL;DR

This paper introduces a task-guided zero-shot domain adaptation method using multi-branch neural networks to learn domain-invariant features, improving transfer performance without target domain data.

Contribution

It proposes a novel end-to-end deep learning approach for ZDA that leverages task guidance and domain invariance, outperforming existing methods.

Findings

Outperforms state-of-the-art ZDA methods on benchmark datasets.

Effectively learns domain-invariant features without target data.

Demonstrates robustness across different tasks and domains.

Abstract

Zero-shot domain adaptation (ZDA) methods aim to transfer knowledge about a task learned in a source domain to a target domain, while data from target domain are not available. In this work, we address learning feature representations which are invariant to and shared among different domains considering task characteristics for ZDA. To this end, we propose a method for task-guided ZDA (TG-ZDA) which employs multi-branch deep neural networks to learn feature representations exploiting their domain invariance and shareability properties. The proposed TG-ZDA models can be trained end-to-end without requiring synthetic tasks and data generated from estimated representations of target domains. The proposed TG-ZDA has been examined using benchmark ZDA tasks on image classification datasets. Experimental results show that our proposed TG-ZDA outperforms state-of-the-art ZDA methods for different domains and tasks.