Representation Learning via Consistent Assignment of Views to Clusters
arXiv:2112.15421 · doi:10.1145/3477314.3507267
Abstract
We introduce Consistent Assignment for Representation Learning (CARL), an unsupervised learning method to learn visual representations by combining ideas from self-supervised contrastive learning and deep clustering. By viewing contrastive learning from a clustering perspective, CARL learns unsupervised representations by learning a set of general prototypes that serve as energy anchors to enforce different views of a given image to be assigned to the same prototype. Unlike contemporary work on contrastive learning with deep clustering, CARL proposes to learn the set of general prototypes in an online fashion, using gradient descent without the necessity of using non-differentiable algorithms or K-Means to solve the cluster assignment problem. CARL surpasses its competitors in many representations learning benchmarks, including linear evaluation, semi-supervised learning, and transfer learning.
arXiv commentPre-print. 37th ACM/SIGAPP Symposium on Applied Computing (SAC'22). Code at https://gitlab.com/mipl/carl/
References in corpus (15)
- A Simple Framework for Contrastive Learning of Visual Representations
- Representation Learning with Contrastive Predictive Coding
- Bootstrap your own latent: A new approach to self-supervised Learning
- Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
- Improved Baselines with Momentum Contrastive Learning
- Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
- Learning deep representations by mutual information estimation and maximization
- Unsupervised Representation Learning by Predicting Image Rotations
- Data-Efficient Image Recognition with Contrastive Predictive Coding
- Learning Representations by Maximizing Mutual Information Across Views
- What Makes for Good Views for Contrastive Learning?
- Large Batch Training of Convolutional Networks
- Prototypical Contrastive Learning of Unsupervised Representations
- Deep Clustering for Unsupervised Learning of Visual Features
- A Downsampled Variant of ImageNet as an Alternative to the CIFAR datasets