CRITICAL ANALYSIS OF SELF-SUPERVISION: Difference between revisions

From statwiki
Jump to navigation Jump to search
Line 3: Line 3:


== Introduction ==
== Introduction ==
This paper evaluated the performance of state-of-the-art  unsupervised (self-supervised)  methods on learning weights of convolutional neural networks (CNNs) to figure out whether current self-supervision techniques can learn deep features from only one image.
The main goal of self-supervised learning is to take advantage of vast amount of unlabeled data for training CNNs and finding  a generalized image representation. 
In self-supervised learning, data generate ground truth labels per se by pretext tasks such as rotation estimation.


== Previous Work ==
== Previous Work ==

Revision as of 00:32, 26 November 2020

Presented by

Maral Rasoolijaberi

Introduction

Previous Work

In recent literature, several papers addressed unsupervised learning methods and learning from a single sample.

A BiGAN [Donahue et al., 2017], or Bidirectional GAN, is basically a generative adversarial network plus an encoder. The generator maps latent samples to generated data and the encoder performs as the opposite of the generator. After training BiGAN, the encoder has learned to generate a rich image representation. In RotNet method [Gidaris et al., 2018], images are rotated and the CNN learns to figure out the direction. DeepCluster [Caron et al., 2018] alternates k-means clustering to learn stable feature representations under several image transformations.

Method

results

Conclusion

Critiques

References

Retrieved from "http://wiki.math.uwaterloo.ca/statwiki/index.php?title=CRITICAL_ANALYSIS_OF_SELF-SUPERVISION&oldid=46584"