The Role of Embedding Complexity in Domain-invariant Representations

Ching-Yao Chuang, Antonio Torralba, Stefanie Jegelka

Keywords: domain adaptation, domain-invariant representations, model complexity, theory, deep learning

TL;DR: We study the effect of the embedding complexity in learning domain-invariant representations and develop a strategy that mitigates sensitivity to it.

Abstract: Unsupervised domain adaptation aims to generalize the hypothesis trained in a source domain to an unlabeled target domain. One popular approach to this problem is to learn domain-invariant embeddings for both domains. In this work, we study, theoretically and empirically, the effect of the embedding complexity on generalization to the target domain. In particular, this complexity affects an upper bound on the target risk; this is reflected in experiments, too. Next, we specify our theoretical framework to multilayer neural networks. As a result, we develop a strategy that mitigates sensitivity to the embedding complexity, and empirically achieves performance on par with or better than the best layer-dependent complexity tradeoff.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:1910.05804/code)