# Neural Networks Questions and Answers – Multiclass Classification

This set of Neural Networks Multiple Choice Questions & Answers (MCQs) focuses on “Neural Networks – Multiclass Classification”.

1. Logistic regression in vanilla form can be used to solve multiclass classification problems.
a) True
b) False

Explanation: Logistic regression is said to be in vanilla form when not customized from their original form, meaning that they are used without any customizations or updates applied to them. In its vanilla form logistic regression is used to do binary classification and not for multiclass classification.

2. Multiclass classification with logistic regression can be done through the one-vs-rest scheme.
a) True
b) False

Explanation: Multiclass classification with logistic regression can be done either through the one-vs-rest scheme in which for each class a binary classification problem of data belonging or not to that class is done, or changing the loss function to cross- entropy loss.

3. The goal of multiclass classification is to construct a function which, given a new data point, will correctly predict the class to which the new point belongs.
a) True
b) False

Explanation: In multiclass classification each training point belongs to one of N different classes. The goal is to construct a function which, given a new data point, will correctly predict the class to which the new point belongs. The points can belong to multiple categories, but a point can belong to only in one category.

4. Softmax layer is one of the best options as the output layer in a multiclass classification implementation.
a) True
b) False

Explanation: While implementing a multiclass classification problem the important part is the choice of the output layer. The usual choice for multi-class classification is the softmax layer. The softmax function is a generalization of the logistic function that “squashes” a KK-dimensional vector z of arbitrary real values to a K – dimensional vector σ (z) of real values in the range [0, 1], that add up to 1.

5. The given neural network cannot be used to solve multiclass classification problems.

a) True
b) False

Explanation: The given figure shows a single input – multi output neural network with two hidden layers. And here it gets input from single source and produce the multiple outputs as in a multiclass classification problem to find out the output class of the given data point. So clearly this neural network can be used for multiclass classification problems.

6. One vs. all becomes increasingly inefficient as the number of classes rises.
a) True
b) False

Explanation: One vs. all involves the training of N-binary classifiers where each of them is designed to recognize a particular class. This approach is fairly reasonable when the total number of classes is small, but becomes increasingly inefficient as the number of classes rises. We can create a significantly more efficient one-vs.-all model with a deep neural network in which each output node represents a different class.

Sanfoundry Global Education & Learning Series – Neural Networks.