Connecting adversarial attacks and optimal transport for domain adaptation

TL;DR

This paper introduces a novel domain adaptation algorithm leveraging optimal transport and adversarial attacks, creating a source fiction domain to improve classifier transfer across domains.

Contribution

The paper proposes a new method that uses c-cyclically monotone transformations and adversarial attacks to enhance domain adaptation via optimal transport.

Findings

Improved classification performance on Digits and Office-31 datasets.

Effective use of simple discrete optimal transport solvers.

Demonstrated the link between adversarial attacks and c-cyclically monotone transformations.

Abstract

We present a novel algorithm for domain adaptation using optimal transport. In domain adaptation, the goal is to adapt a classifier trained on the source domain samples to the target domain. In our method, we use optimal transport to map target samples to the domain named source fiction. This domain differs from the source but is accurately classified by the source domain classifier. Our main idea is to generate a source fiction by c-cyclically monotone transformation over the target domain. If samples with the same labels in two domains are c-cyclically monotone, the optimal transport map between these domains preserves the class-wise structure, which is the main goal of domain adaptation. To generate a source fiction domain, we propose an algorithm that is based on our finding that adversarial attacks are a c-cyclically monotone transformation of the dataset. We conduct experiments on Digits and Modern Office-31 datasets and achieve improvement in performance for simple discrete optimal transport solvers for all adaptation tasks.