Skip to content

πŸ«πŸ™‹πŸ» [WIP] Self learn Federated Learning (FL). This repository is a collection of resources, notebooks, blogs, and tutorials for beginners.

License

Notifications You must be signed in to change notification settings

snehilsanyal/self-learn-federated-learning

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

6 Commits
Β 
Β 
Β 
Β 

Repository files navigation

πŸ«πŸ™‹πŸ» Self Learn Federated Learning

βš’ πŸ’» Work in Progress

This repository contains a roadmap for beginner's to start their journey on Federated Learning. Self-curated. The difficulty will be moderate and expects prior knowledge about Python and Deep Learning concepts.

πŸ‘‹πŸ» Introductory Resources

  1. Start off with this funny and insightful comic on Federated Learning by Google AI. The same site contains a list of learning resources and research done by Google on Federated Learning.

    Link: Federated Learning Online Comic

  2. The first paper on Federated learning was by Google Inc in 2017. The authors presented a new learning paradigm where the data remains distributed in several mobile devices but the model is trained in a decentralized way. Instead of data moving to a centralized server, the model moves to distributed devices. Quoting from the Abstract (had to mention it lol):

    We advocate an alternative that leaves the training data distributed on the mobile devices, and learns a shared model by aggregating locally-computed updates. We term this decentralized approach Federated Learning.

    Blog: Federated Learning: Collaborative Machine Learning without Centralized Training Data

    Paper: Communication-efficient learning of deep networks from decentralized data

    Read the blog first, to get an overall picture of the work, then jump on to the paper for the nitty-gritty details.

  3. The next paper is again by Google Research. This time the paper deals with Keyword Prediction

βš’πŸ’» Hands-Dirty FL

πŸ—ƒπŸ’» FL Libraries

In the order of ease of usage (currently trying Flower, Syft and TF-Federated, will update the order):

  1. Flower by Flower Labs
  2. TensorFlow Federated by Google PARFAIT
  3. PySyft by OpenMined
  4. FATE by FedAI Ecosystem
  5. CLARA by NVIDIA
  6. Substra by Owkin

πŸ€·πŸ»β€β™‚οΈπŸ’» FL Tutorials

FL Usecases

πŸ“°πŸ“œ FL Papers

πŸ“šπŸ“” FL Books

FL Courses

  1. [DL.AI X Flower Labs, Short Course on Intro to Federated Learning]
  2. [DL.AI X Flower Labs, Short Course on Federated Fine-tuning of LLMs with Private Data]

FL Blogs

FL Research Labs

  1. Federated GitHub by Google Research:A collection of Google research projects related to Federated Learning and Federated Analytics.

FL Communities

  1. The Federated Learning Portal: This portal keeps track of books, workshops, conferences, special tracks, and other events related to the field of FL. I came to know about many competitions in the domain of FL from this webpage.
  2. OpenMined Slack
  3. Flower Labs Slack
  4. FedML Discord

πŸ±β€πŸ’»πŸ•Ή FL Hackathons, Competitions and Challenges

πŸ’πŸ› FL Organizations

  1. OpenMined

FL Companies

  1. Flower Labs
  2. Owkin

About

πŸ«πŸ™‹πŸ» [WIP] Self learn Federated Learning (FL). This repository is a collection of resources, notebooks, blogs, and tutorials for beginners.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published