Controlling Directions Orthogonal to a Classifier
Abstract
We propose to identify directions invariant to a given classifier so that these directions can be controlled in tasks such as style transfer. While orthogonal decomposition is directly identifiable when the given classifier is linear, we formally define a notion of orthogonality in the non-linear case. We also provide a surprisingly simple method for constructing the orthogonal classifier (a classifier utilizing directions other than those of the given classifier). Empirically, we present three use cases where controlling orthogonal variation is important: style transfer, domain adaptation, and fairness. The orthogonal classifier enables desired style transfer when domains vary in multiple aspects, improves domain adaptation with label shifts and mitigates the unfairness as a predictor. The code is available at http://github.com/Newbeeer/orthogonal_classifier
Cite
@article{arxiv.2201.11259,
title = {Controlling Directions Orthogonal to a Classifier},
author = {Yilun Xu and Hao He and Tianxiao Shen and Tommi Jaakkola},
journal= {arXiv preprint arXiv:2201.11259},
year = {2022}
}
Comments
accepted by ICLR 2022