🛰️
SIT-FUSE
  • 🛰️SIT-FUSE Docs
    • Machine Learning Approach
    • 🤖Self-Supervised Representation Learning
      • RBMs and DBNs
      • BYOL
      • I-JEPA
    • 🔲Self-Supervised Segmentation
  • User Guide
    • Installing the repo
    • Code Configuration
      • Training and Dataset Generation
        • Config Options
          • Encoder Config Options
            • Encoder Training Config Options
              • Training data Stratification Config Options
          • DBN-Specific Config Options
          • BYOL-Specific Config Options
          • I-JEPA-Specific Training Options
          • Deep Clustering Config Options
            • Deep Clustering Training Config Options
            • Hierarchical Deep Clustering Config Options
              • Hierarchichal Deep Clustering Training Config Options
          • Logger Config Options
          • Scaler Config Options
          • Output Config Options
          • Data Config Options
            • Reader Options
        • Example Configs
          • Example DBN Config
          • Example BYOL Config
          • Example I-JEPA Config
      • Geolocation Application
        • Config Options
          • Data Config Options
          • Context Config Options
        • Example Config
      • Data combination, colocation, and resampling
        • Config Options
          • Low Res Config Options
            • Data Config Options
          • High Res Config Options
            • Data Config Options
          • Fusion Config Options
        • Example Config
    • Dockerized Environment
    • Conda Virtual Environment
    • Training
      • Training Dataset Preprocessing Examples
      • Training Command Examples
      • Inference Command Examples
      • Training Summary
    • File Conversion
    • Post-processing
      • Labeling Tutorial
        • QGIS configuration
        • Understanding the data
        • Creating and modifying shapefiles
        • Other shapefile tools
      • Zonal Histogram
        • Creating the YAML file
        • Gathering files
        • Code configuration
        • Getting output
      • Context Assignment
        • Class comparison
        • GeoTiff generation
        • Contours
  • SIT FUSE Comparison
    • Precise Output
Powered by GitBook
On this page
  1. SIT-FUSE Docs
  2. Self-Supervised Representation Learning

BYOL

Bootstrap Your Own Latent

PreviousRBMs and DBNsNextI-JEPA

Last updated 8 months ago

Bootstrap Your Own Latent - A framework that tries to minimize similarity loss between a pair of convolutional networks trained together, one seeing the unperturbed version of a sample, and the other seeing its perturbed pair

References:

  1. Jean-Bastien Grill, Florian Strub, Florent Altché, Corentin Tallec, Pierre H. Richemond, Elena Buchatskaya, Carl Doersch, Bernardo Avila Pires, Zhaohan Daniel Guo, Mohammad Gheshlaghi Azar, Bilal Piot, Koray Kavukcuoglu, Rémi Munos, Michal Valko: “Bootstrap your own latent: A new approach to self-supervised Learning”, 2020; arXiv:2006.07733.

🛰️
🤖