https://github.com/the-ai-summer/byol-cifar10

implement byol in cifar-10
https://github.com/the-ai-summer/byol-cifar10

Last synced: 2 months ago
JSON representation

implement byol in cifar-10

• Host: GitHub
• URL: https://github.com/the-ai-summer/byol-cifar10
• Owner: The-AI-Summer
• License: mit
• Created: 2022-04-07T08:20:23.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2022-05-09T12:29:18.000Z (over 3 years ago)
• Last Synced: 2025-07-09T14:47:29.867Z (3 months ago)
• Language: Jupyter Notebook
• Size: 67.4 KB
• Stars: 15
• Watchers: 2
• Forks: 3
• Open Issues: 3
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# BYOL tutorial: self-supervised learning on CIFAR images with code in Pytorch

An education step by step implementation of BYOL that accompanies the [blogpost](https://theaisummer.com/byol/)

After presenting SimCLR, a contrastive self-supervised learning framework, I decided to demonstrate another infamous method, called BYOL. Bootstrap Your Own Latent (BYOL), is a new algorithm for self-supervised learning of image representations. BYOL has two main advantages: 

It does not explicitly use negative samples. Instead, it directly minimizes the similarity of representations of the same image under a different augmented view (positive pair). Negative samples are images from the batch other than the positive pair.

As a result, BYOL is claimed to require smaller batch sizes, which makes it an attractive choice.

[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/16Mjujx6aLZX7wcge_2Xca0NdQyrONPA1?usp=sharing)