Deep Learning · End Term · 21 Dec 2025 · September 2025 term

Question 13: Why is the sigmoid activation function often preferred o…

Question 13

+3 marksOne or more correct options

Why is the sigmoid activation function often preferred over the step activation function in neural networks?

Select all that apply.

A
The sigmoid is differentiable, allowing gradient-based optimization methods like backpropagation.
B
The step function is non-differentiable and does not support learning through gradient descent.
C
The sigmoid function completely avoids vanishing gradient problems.
D
The step function outputs values in the range (0,1), whereas the sigmoid outputs only discrete values {0,1}.

Show answer

Correct answers

A
The sigmoid is differentiable, allowing gradient-based optimization methods like backpropagation.
B
The step function is non-differentiable and does not support learning through gradient descent.

See the whole paper Take it as a mock test

Question 13 of 32 in the IIT Madras BS Deep Learning (Deep Learning) End Term paper sat on 21 Dec 2025, in the September 2025 term (Deep Learning 18 Dec 25). It carries 3 marks.

Question 13: Why is the sigmoid activation function often preferred o…

Question 13

More questions from this paper