Recent Advancements in Self-Supervised Paradigms for Visual Feature Representation

Abstract

We witnessed a massive growth in the supervised learning paradigm in the past decade. Supervised learning requires a large amount of labeled data to reach state-of-the-art performance. However, labeling the samples requires a lot of human annotation. To avoid the cost of labeling data, self-supervised methods were proposed to make use of largely available unlabeled data. This study conducts a comprehensive and insightful survey and analysis of recent developments in the self-supervised paradigm for feature representation. In this paper, we investigate the factors affecting the usefulness of self-supervision under different settings. We present some of the key insights concerning two different approaches in self-supervision, generative and contrastive methods. We also investigate the limitations of supervised adversarial training and how self-supervision can help overcome those limitations. We then move on to discuss the limitations and challenges in effectively using self-supervision for visual tasks. Finally, we highlight some open problems and point out future research directions.

References in corpus (17)

• A Simple Framework for Contrastive Learning of Visual Representations
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Deep Residual Learning for Image Recognition
• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Neural Discrete Representation Learning
• Improved Baselines with Momentum Contrastive Learning
• Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
• Conditional Image Generation with PixelCNN Decoders
• Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples
• Adversarial Feature Learning
• What Makes for Good Views for Contrastive Learning?
• Contrastive Multiview Coding
• Adversarial Machine Learning at Scale
• On Mutual Information Maximization for Representation Learning
• Understanding the Limitations of Variational Mutual Information Estimators
• How Useful is Self-Supervised Pretraining for Visual Tasks?