Python for Deep Learning - Build Neural Networks in Python - Hyperbolic Tangent Function

The video tutorial explains the hyperbolic tangent function, comparing it to the sigmoid function. It highlights the advantages of the hyperbolic tangent function, such as its ability to map negative inputs strongly negative and its use in classification tasks. The function is particularly useful in feedforward propagation to enhance output differences.

What is the range of the hyperbolic tangent function?

• -1 to 0

• 0 to 2

• -1 to 1

• 0 to 1

Why is the hyperbolic tangent function considered advantageous over the sigmoid function?

• It has a larger range

• It is not monotonic

• It maps negative inputs strongly negative

• It is not differentiable

How does the hyperbolic tangent function handle zero inputs?

• Maps them to 1

• Maps them to 0.5

• Maps them to -1

• Maps them near 0

In what type of tasks is the hyperbolic tangent function mainly used?

• Dimensionality reduction

• Regression tasks

• Classification between two classes

• Clustering tasks

What is the role of the hyperbolic tangent function in feedforward propagation?

• To eliminate negative outputs

• To maintain the same output

• To increase the difference between outputs

• To decrease the difference between outputs