A Lightweight Hybrid Deep Learning Privacy Preserving Model for FC-Based Industrial Internet of Medical Things

. 2022 Mar 9;22(6):2112. doi: 10.3390/s22062112

Algorithm 1: FC-Average Algorithm

1: Init: a = 0;
2: for each round t = 1,2,…do
3: select K clients
4: for each K clients do
5: w^kt, UpdateClient( )
6: d^kt ← the distance between two classes dataset end for
7: If (d^k = t) ←1
8: d^k = w^k * t ←p^k k = 1nk f(d^kt) wt1/pkk = 1nkf^d kt)
9: end for
10: Updatefunction( )
11: Initialize local minibatch size L, local epochs E, learning rate
12: for each epoch i E do
13: randomly choose S: based on size L
14: wi ← w1 – w5g(w1:s)
15: end for
16: return i
17: End Procedure
18: End Algorithm