This repository contains the code to run simulations from the 'DPFed: Toward Fair Personalized Federated Learning with Fast Convergence'.
Contains implementations of FedAvg, MTFL [1], Per-FedAvg [2] and pFedMe [3] as described in the paper.
Package | Version |
---|---|
python | 3.8 |
pytorch | 1.7.0 |
torchvision | 0.8.1 |
numpy | 1.21.3 |
progressbar2 | 3.47.0 |
Requires Fashion-MNIST and CIFAR10.
Run main.py. Each experiment setting requires different command-line arguments. Will save a .pkl
file in the 'result' directory containing experiment data as numpy arrays.
[1] Multi-Task Federated Learning for Personalised Deep Neural Networks in Edge Computing, Mills et al. IEEE TPDS 2022.
[2] Personalized Federated Learning with Theoretical Guarantees: A Model-Agnostic Meta-Learning Approach, Fallah et al. NeurIPS 2020.
[3] Personalized Federated Learning with Moreau Envelopes, Dinh et al. NeurIPS 2020.
[4] Addressing Function Approximation Error in Actor-Critic Methods, Fujimoto et al. ICML 2018.