Documentation

Federated Learning (FL) [1,2,3] is a learning paradigm for collaboratively learning models from isolated data without directly sharing privacy information, which helps to satisfy the requirements of privacy protection of the public. FederatedScope (FS), a comprehensive federated learning platform with event-driven architecture, aims to provide easy-to-use and flexible support for users who want to get started and customize task-specific FL procedures quickly.

Tutorials

• For beginner

  • Installation

  • Start With Examples

  • Start Your Own Case

  • DataZoo

  • ModelZoo

  • AlgoZoo

  • Tuning Federated Learning

• For advanced users or developer

  • FS Data module

  • Event-driven Architecture

  • New Types of Messages and Handlers

  • Privacy Protection for Message

  • Local Learning Abstraction: Trainer

  • Personalized FL

  • Cross-Device FL

  • Cross-Silo FL

  • Accelerating Federated HPO

  • Simulation and Deployment

• Application with FS

  • Recommendation

  • Differential Privacy

  • Privacy Attacks

  • Graph

  • NLP and Speech

  • LLM

• Get involved

  • Join Our Community

  • Contributing to FederatedScope

API References

• Core Module References
• Federated Computer Vision Module References
• Federated Natural Language Processing Module References
• Federated Graph Learning Module References
• Auto-tuning Module References
• Attack Module References
• Federated Matrix Factorization Module References

References

[1] McMahan B, Moore E, Ramage D, et al. “Communication-efficient learning of deep networks from decentralized data”. Artificial intelligence and statistics. PMLR, 2017: 1273-1282. [2] Konečný J, McMahan H B, Ramage D, et al. “Federated optimization: Distributed machine learning for on-device intelligence”. arXiv preprint arXiv:1610.02527, 2016.
[3] Yang Q, Liu Y, Cheng Y, et al. “Federated learning”. Synthesis Lectures on Artificial Intelligence and Machine Learning, 2019, 13(3): 1-207.