What function do nonlinear activation functions like sigmoid, tanh$,$ and ReLU serve within neural networks?

Question $5$Answer

a$.$

They speed up the gradient calculation in backpropagation, as compared to linear units

b$.$

The nonlinear activation functions are differentiable to update the weights

c$.$

They help to learn nonlinear decision boundaries

d$.$

The nonlinear activation functions prevent the vanishing gradient problem

e$.$

They always output values between $0$ and $1$

f$.$

None