Self-Supervised Learning Disentangled Group Representation as Feature

Tan Wang, Zhongqi Yue, Jianqiang Huang, Qianru Sun, Hanwang Zhang

Keywords: Self-Supervised Learning, Feature Disentanglement, Invariant Risk Minimization, Group Representation

TL;DR: An iterative IRM algorithm for unsupervised feature disentanglement and self-supervised feature learning

Abstract: A good visual representation is an inference map from observations (images) to features (vectors) that faithfully reflects the hidden modularized generative factors (semantics). In this paper, we formulate the notion of "good" representation from a group-theoretic view using Higgins' definition of disentangled representation, and show that existing Self-Supervised Learning (SSL) only disentangles simple augmentation features such as rotation and colorization, thus unable to modularize the remaining semantics. To break the limitation, we propose an iterative SSL algorithm: Iterative Partition-based Invariant Risk Minimization (IP-IRM), which successfully grounds the abstract semantics and the group acting on them into concrete contrastive learning. At each iteration, IP-IRM first partitions the training samples into two subsets that correspond to an entangled group element. Then, it minimizes a subset-invariant contrastive loss, where the invariance guarantees to disentangle the group element. We prove that IP-IRM converges to a fully disentangled representation and show its effectiveness on various benchmarks. Codes are available at https://github.com/Wangt-CN/IP-IRM.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

Code: https://github.com/Wangt-CN/IP-IRM

Community Implementations: [ 10 code implementations](https://www.catalyzex.com/paper/arxiv:2110.15255/code)