You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Maybe give motivations about why the paper should be implemented as a baseline.
Floco is an effective method for improving personalized model performance in the non-IID cross-silo federated learning setting compared to many state-of-the-art personalized FL methods while increasing global model performance compared to FedAvg and FedProx. It achieves this by training a shared parameter simplex, called solution simplex, across clients. This approach assigns similar clients to similar regions within the simplex, leading to better collaboration of similar clients, i.e. clients with similar data distributions, and less interference with dissimilar clients.
The goal of this issue is to reproduce the SimpleCNN CIFAR-10 results from the original paper. The resulting code can be easily extended to the full experimental setup of the paper.
Is there something else you want to add?
No response
Implementation
To implement this baseline, it is recommended to do the following items in that order:
Paper
Dennis Grinwald, Philipp Wiesner, Shinichi Nakajima. Federated Learning over Connected Modes (NeurIPS'25)
Link
https://openreview.net/pdf?id=JL2eMCfDW8
Maybe give motivations about why the paper should be implemented as a baseline.
Floco is an effective method for improving personalized model performance in the non-IID cross-silo federated learning setting compared to many state-of-the-art personalized FL methods while increasing global model performance compared to FedAvg and FedProx. It achieves this by training a shared parameter simplex, called solution simplex, across clients. This approach assigns similar clients to similar regions within the simplex, leading to better collaboration of similar clients, i.e. clients with similar data distributions, and less interference with dissimilar clients.
The goal of this issue is to reproduce the SimpleCNN CIFAR-10 results from the original paper. The resulting code can be easily extended to the full experimental setup of the paper.
Is there something else you want to add?
No response
Implementation
To implement this baseline, it is recommended to do the following items in that order:
For first time contributors
first contribution
docPrepare - understand the scope
Verify your implementation
EXTENDED_README.md
that was created in your baseline directoryREADME.md
is ready to be run by someone that is no familiar with your code. Are all step-by-step instructions clear?README.md
and verify everything runs.The text was updated successfully, but these errors were encountered: