Table 2.

Summary of the experiments.

Name	Learning strategy
A	Centralized training using stochastic gradient descent (learning rate: 0.9)
B	Collaborative training of 5 workers using the round robin method (learning rate: 0.9)
C_0.1	Collaborative privacy preserving training of 5 workers with DSSGDa (θd = 0.1, θu = 0.5, γ = 10, τ = 0.0001; learning rate: 0.9)
C_0.5	Collaborative privacy preserving training of 5 workers with DSSGD (θd = 0.5, θu = 0.5, γ = 10, τ = 0.0001; learning rate: 0.9)
D	Local training without collaboration of 5 workers using stochastic gradient descent

^aDSSGD: distributed selective stochastic gradient descent.