Stochastic Contrastive Learning

TL;DR

This paper introduces latent variable approximations into large-scale contrastive self-supervised learning models, enhancing performance, interpretability, and compression of learned representations.

Contribution

It is the first to integrate latent variable models with contrastive SSL, improving downstream results and interpretability.

Findings

Achieved 96.42% test accuracy on CIFAR10

Achieved 77.49% test accuracy on ImageNet

Produced 588x compressed representations

Abstract

While state-of-the-art contrastive Self-Supervised Learning (SSL) models produce results competitive with their supervised counterparts, they lack the ability to infer latent variables. In contrast, prescribed latent variable (LV) models enable attributing uncertainty, inducing task specific compression, and in general allow for more interpretable representations. In this work, we introduce LV approximations to large scale contrastive SSL models. We demonstrate that this addition improves downstream performance (resulting in 96.42% and 77.49% test top-1 fine-tuned performance on CIFAR10 and ImageNet respectively with a ResNet50) as well as producing highly compressed representations (588x reduction) that are useful for interpretability, classification and regression downstream tasks.