Deep Learning Quiz

What is the key to deep learning?

• Application of decision trees

• Use of linear regression

• Creation of an artificial neural network

• Implementation of support vector machines

What are artificial neural networks made up of?

• Single layer of neurons

• Multiple layers of neurons

• Randomly connected neurons

• No neurons at all

What kind of machine learning problem can deep neural networks be used for?

• Only regression problems

• None of the above

• Only classification problems

• Regression and classification problems

What is the process of inferencing a predicted penguin class using a neural network?

• Assigning random labels to the data

• Using a decision tree for prediction

• Feeding the data through the network and calculating output values

• Ignoring the input data completely

What is the purpose of a loss function in neural network training?

• To optimize the weights in the network

• To compare the predicted values to the known values and aggregate the difference

• To calculate the accuracy of the model

• To determine the number of layers in the network

What technique is used to evaluate the influence of each weight in the network on the loss?

• Trial and error method

• Brute force method

• Random guessing

• Gradient descent approach

What is backpropagation in the context of neural network training?

• Ignoring the loss function

• Feeding the data through the network in reverse order

• Replacing the previously used weight values

• Optimizing the activation functions

What is the purpose of batch processing in neural network training?

• To optimize the weights in the network

• To process the data one at a time

• To speed up the training process

• To perform linear algebraic calculations

What kind of processing units are best for neural network training?

• Solid state drives (SSDs)

• Random access memory (RAM)

• Graphics processing units (GPUs)

• Central processing units (CPUs)

What is the main goal of the training process in deep learning?

• To minimize the number of layers in the network

• To maximize the loss function

• To learn the weights that will result in the most accurate predictions

• To ignore the input data completely