A Two-Step Disentanglement Method

Naama Hadad, Lior Wolf, Moni Shahar; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018, pp. 772-780


We address the problem of disentanglement of factors that generate a given data into those that are correlated with the labeling and those that are not. Our solution is simpler than previous solutions and employs adversarial training. First, the part of the data that is correlated with the labels is extracted by training a classifier. Then, the other part is extracted such that it enables the reconstruction of the original data but does not contain label information. The utility of the new method is demonstrated on visual datasets as well as on financial data. Our code is available at https://github.com/naamahadad/A-Two-Step-Disentanglement-Method.

Related Material

author = {Hadad, Naama and Wolf, Lior and Shahar, Moni},
title = {A Two-Step Disentanglement Method},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2018}