Breaching privacy in federated learning scenarios for vision and text

DISCO is a code-free and installation-free browser platform that allows any non-technical user to collaboratively train machine learning models without sharing any private data.

The official implementation of the paper "Topology-aware Generalization of Decentralized SGD"

[ICML 2023] Decentralized SGD and Average-direction SAM are Asymptotically Equivalent

[IEEE TSIPN' 2022] "Scalable Perception-Action-Communication Loops with Convolutional and Graph Neural Networks", by Ting-Kuei Hu, Fernando Gama, Tianlong Chen, Wenqing Zheng, Zhangyang Wang, Alejandro Ribeiro, and Brian M. Sadler

source code of AISTATS 2020 paper: Fully Decentralized Joint Learning of Personalized Models and Collaboration Graphs

[TMLR] CoDeC: Communication-Efficient Decentralized Continual Learning

[CVPR 2023] CaPriDe Learning: Confidential and Private Decentralized Learning based on Encryption-friendly Distillation Loss

Investigating the reproducibility of federated GNN models

code for "Beyond Exponential Graph: Communication-Efficient Topologies for Decentralized Learning via Finite-time Convergence"

Device-to-Device (D2D) and assocaited Federated Learning simulation in Python using Pytorch for Learning operations. Implemented simultaneous operation of multiple FL modes.

subMFL: Compatible subModel Generation for Federated Learning in Device Heterogeneous Environment

A blockchain-based ecosystem for Open Educational Resources (OERs) built on Ethereum. This project aims to establish standards and infrastructure for creating, sharing, and managing educational content using blockchain technology.

Decentralized and Privacy-Preserving Machine Learning: Exploring the Power of Federated Learning.

pytorch implementation of Optimization as a Model for Few-shot Learning

Handwritten Digit Classification using Federated Learning