Table - PMC

Skip to main content

View full-text article in PMC

. 2024 Dec 23;14(24):2891. doi: 10.3390/diagnostics14242891

Algorithm 1 The FedAvg algorithm. K clients are indexed by k, B is the local minibatch size, E is the number of local epochs, and

η

is the learning rate.

1:
Server executes:
2:
initialize $w_{0}$
3:
for $each round t = 1, 2, \dots$ do
4:
$m \leftarrow max (C \cdot K, 1)$
5:
$S_{t} \leftarrow$ (random set of m clients)
6:
for $each client k \in S_{t} in parallel$ do
7:
$w_{t + 1}^{k} \leftarrow ClientUpdate (k, w_{t})$
8:
end for
9:
$w_{t + 1} \leftarrow \sum_{k = 1}^{K} \frac{n_{k}}{n} w_{t + 1}^{k}$
10:
end for
11:
ClientUpdate $(k, w)$ : //run on client k
12:
$B \leftarrow$ (split $P_{k}$ into batches of size B)
13:
for $each local epoch i from 1 to E$ do
14:
for $each batch b \in B$ do
15:
$w \leftarrow w - η \cdot \nabla l (w, b)$
16:
end for
17:
end for
18:
return w to Server