461 views
0 0 votes

Consider a neural network with the number of neurons in different layers as mentioned in the below list for a regression task. The first layer is the input layer and the last is the output layer.

$$
[5,5,4,3,1]
$$


What will be an appropriate activation function for the output layer?

  1. SIGMOID
     
  2. LINEAR
     
  3. RELU
     
  4. TANH

1 Answer

0 0 votes
Sigmoid is used for binary classification at the output layer.

ReLU is mostly used at the hidden layers. (It does not make sense to be used at the output layer.)
Same goes for tanh

Linear is an appropriate activation function for the output layer in a regression problem. Typically, this is just the identity function; that is, $g(z)=z$.
Answer:
Position:
Show:

Related questions

1 1 vote
1 1 answer
735
735 views
GO Classes asked Oct 13, 2025
735 views
Suppose we build a neural network for a 5-class classification task. Suppose for a single training example, the true label is $\left[\begin{array}{lllll}1 & 0 & 0 & 0 & 0...
1 1 vote
1 1 answer
477
477 views
GO Classes asked Oct 13, 2025
477 views
Consider a neural network with the number of neurons in different layers as mentioned in the below list for a regression task. The first layer is the input layer and the ...
0 0 votes
1 1 answer
428
428 views
GO Classes asked Oct 13, 2025
428 views
Consider a neural network with the number of neurons in different layers as mentioned in the below list for a regression task. The first layer is the input layer and the ...
3 3 votes
1 1 answer
538
538 views
GO Classes asked Oct 10, 2025
538 views
Consider the following data:\[\begin{array}{|l|l|}\hlineg(x) & y \\\hline+30 & +1 \\\hline-20 & -1 \\\hline-1 & -1 \\\hline+1 & +1 \\\hline\end{array}\]Which of the follo...