A Survey on Contrastive Self-supervised Learning

TL;DR

This paper reviews contrastive self-supervised learning, highlighting its methods, architectures, and performance across various domains, while discussing current limitations and future research directions.

Contribution

It provides a comprehensive survey of contrastive self-supervised learning methods, architectures, and their performance, along with insights into limitations and future challenges.

Findings

Contrastive methods perform well on multiple downstream tasks.

Different architectures show varying effectiveness across domains.

Current methods have notable limitations that require further research.

Abstract

Self-supervised learning has gained popularity because of its ability to avoid the cost of annotating large-scale datasets. It is capable of adopting self-defined pseudo labels as supervision and use the learned representations for several downstream tasks. Specifically, contrastive learning has recently become a dominant component in self-supervised learning methods for computer vision, natural language processing (NLP), and other domains. It aims at embedding augmented versions of the same sample close to each other while trying to push away embeddings from different samples. This paper provides an extensive review of self-supervised methods that follow the contrastive approach. The work explains commonly used pretext tasks in a contrastive learning setup, followed by different architectures that have been proposed so far. Next, we have a performance comparison of different methods for multiple downstream tasks such as image classification, object detection, and action recognition. Finally, we conclude with the limitations of the current methods and the need for further techniques and future directions to make substantial progress.