SimCLR: A Simple Framework for Contrastive Learning of Visual Representations

Abstract

This paper presents SimCLR, a simple framework for contrastive learning of visual representations. We simplify recently proposed contrastive self-supervised learning algorithms without requiring specialized architectures or a memory bank. In order to understand what enables the contrastive prediction tasks to learn useful representations, we systematically study the major components of our framework. We show that (1) composition of data augmentations plays a critical role in defining effective predictive tasks, (2) introducing a learnable nonlinear transformation between the representation and the contrastive loss substantially improves the quality of the learned representations, and (3) contrastive learning benefits from larger batch sizes and more training steps compared to supervised learning. By combining these findings, we are able to considerably outperform previous methods for self-supervised and semi-supervised learning on ImageNet. A linear classifier trained on self-supervised representations learned by SimCLR achieves 76.5% top-1 accuracy, which is a 7% relative improvement over previous state-of-the-art, matching the performance of a supervised ResNet-50. When fine-tuned on only 1% of the labels, we achieve 85.8% top-5 accuracy, outperforming AlexNet with 100X fewer labels.

Date
Nov 1, 2023 1:30 PM — 2:00 PM
Event
EMIL Fall'23 Seminars
Location
Health Futures Center, ASU
Shovito Barua Soumma
Shovito Barua Soumma
Graduate Research Associate
PhD Student

Currently I am working on building and optimizing deep learning models for wearable sensors data.