Artificial Intelligence MCQ - Artificial Intelligence Section 2

71 What should be the loss function used to train the model given in the above question?

72 A Convolutional Neural Network (CNN) for an image classification task has the following operations.
1. Max Pooling
2. Convolution Operation
3. Flatten
4. Fully Connected Layer
Identify the correct sequence from the options below:

73 A vanilla autoencoder is a Neural Network architecture used to create lower dimensional input representation. Which of the following statements re true about it?
1. It is an unsupervised algorithm
2. It can generate new data by learning the probability distribution
3. Its target output is the input
4. Autoencoders have linear encoder and decoder

Consider a two-layer network of the form shown in Figure 1 with M hidden units having Φ(.) = tanh(.) activation functions and full connectivity in both layers.

Figure1: Network diagram for the two-layer neural network corresponding to questions 1,2 and 3. The input, hidden, and output variables are represented by nodes, and the weight parameters are represented by links between the nodes, in which the bias parameters are denoted by links coming from additional input and hidden variables x₀ and z₀. Arrows denote the direction of information flow through the network during forward propagation

74 One property of feed-forward network shown in Figure 1 is that multiple distinct choices for the weight vector w can all give rise to the mapping function from inputs to outputs.

75 Let x and y be two unit norm vectors. Then minimizing the Euclidean distance between x and y corresponds to minimizing the inner product of x and y