-
https://github.com/LeiDu-dev/FedSGD/blob/main/model/server.py#L37-L38
-
Key points:
- Cum Laude ambitions
- thesis _science_ must come above everything else
- possible direction with most science imaginable
- TBM compatible; thus governance of the global financial sys…
-
Very glad to see your well done work, but there is a private question for you.
When you are using FEMNIST dataset, did you sample or use the full dataset?
-
Hello! I'm trying to reproduct your paper, but fail in the case of non-iid data on cifar10. I just input command like you recommend:
python train_classifier_fed.py --data_name CIFAR10 --model_name …
-
This method is called (on wikipedia), [Federated Stochastic Gradient Descent (FedSGD)](https://en.wikipedia.org/wiki/Federated_learning#Federated_Stochastic_Gradient_Descent_(FedSGD)).
I guess this…
-
https://arxiv.org/pdf/1602.05629.pdf
一开始我以为是异步,结果发现是同步....不过是delayed同步. 本地更新几步然后再sync
-
Dear author, thanks for sharing your wonderful codes.
In the ICLR paper Algorithm 1, the server updates w by (sum of deltas) / (sum of hs). As I understand from the paper, the h_k^t's are the upper…
-
**Is your feature request related to a problem? Please describe.**
Right now the federated optimizer API assumes that we are only going to use the same optimizer across different workers.
**Descri…