ML Quiz 14

Is logistic regression for multi-class possible?

• yes

• no

Which activation function is used mostly for multi-class in output layer?

• Relu

• Softmax

• Sigmoid

• Adam

Which activation function is used mostly for multi-class in input layer?

• Relu

• Softmax

• Sigmoid

• Adam

Which activation function is used for non-linear problem in binary classification ?

• Relu

• Softmax

• Sigmoid

• Adam

Which loss function is used in binary classification?

• binary class entropy

• categorical cross entropy

___________ is used to update weight for backpropagation.

• Loss function

• Optimizer

• Activation function

If we train CNN with less images, then there is a chance of _______.

• Overfitting

• Underfitting

_______ calculates for all the positive labels, but only for a random sample of negatives.

• Candidate Sampling

• Full SoftMax

When the number of classes is large enough, we can use _______

• candidate sampling

• full softmax

Which can help prevent vanishing gradients?

• ReLU

• Sigmoid

• Softmax

• Adam