Adversarial examples are transferable between different models. In our paper, we propose to use this property for multi-step domain adaptation. In unsupervised domain adaptation settings, we demonstrate that replacing the source domain with adversarial examples to HΔH-divergence can improve source classifier accuracy on the target domain. Our method can be connected to most domain adaptation techniques. We conducted a range of experiments and achieved improvement in accuracy on Digits and Office-Home datasets.
@article{arxiv.2207.08948,
title = {Multi-step domain adaptation by adversarial attack to $\mathcal{H} \Delta \mathcal{H}$-divergence},
author = {Arip Asadulaev and Alexander Panfilov and Andrey Filchenkov},
journal= {arXiv preprint arXiv:2207.08948},
year = {2022}
}