Transductive Adversarial Networks (TAN)

TL;DR

Transductive Adversarial Networks (TAN) is a flexible domain-adaptation framework that uses adversarial training to learn conditional distributions on unlabelled target data, accommodating different tasks across domains.

Contribution

TAN introduces a novel adversarial architecture for domain adaptation that handles different tasks in source and target domains, with theoretical validation.

Findings

Handles different source and target tasks.

Uses adversarial training for complex distributions.

Theoretically validated framework.

Abstract

Transductive Adversarial Networks (TAN) is a novel domain-adaptation machine learning framework that is designed for learning a conditional probability distribution on unlabelled input data in a target domain, while also only having access to: (1) easily obtained labelled data from a related source domain, which may have a different conditional probability distribution than the target domain, and (2) a marginalised prior distribution on the labels for the target domain. TAN leverages a fully adversarial training procedure and a unique generator/encoder architecture which approximates the transductive combination of the available source- and target-domain data. A benefit of TAN is that it allows the distance between the source- and target-domain label-vector marginal probability distributions to be greater than 0 (i.e. different tasks across the source and target domains) whereas other domain-adaptation algorithms require this distance to equal 0 (i.e. a single task across the source and target domains). TAN can, however, still handle the latter case and is a more generalised approach to this case. Another benefit of TAN is that due to being a fully adversarial algorithm, it has the potential to accurately approximate highly complex distributions. Theoretical analysis demonstrates the viability of the TAN framework.