Deep Learning Fundamentals

What is a neural network?

• A neural network is a social network for connecting people.

• A neural network is a biological system found in plants.

• A neural network is a type of hardware used for gaming.

• A neural network is a computational model that simulates the way human brains process information, consisting of interconnected layers of nodes.

What is the purpose of an activation function?

• The purpose of an activation function is to introduce non-linearity into the neural network.

• To normalize the input data before processing.

• To reduce the learning rate during training.

• To increase the number of layers in the network.

How do you evaluate the performance of a neural network?

• Evaluate based on the number of neurons in each layer.

• Count the number of layers in the network.

• Use metrics like accuracy, precision, recall, F1 score, and loss on validation datasets.

• Measure the training time only.

What are the main layers in a neural network?

• Input layer, Output layer, Dropout layer

• Input layer, Hidden layer(s), Output layer

• Input layer, Hidden layer, Activation layer

• Input layer, Output layer, Bias layer

What is the role of the input layer in a neural network?

• The input layer is responsible for training the model.

• The input layer generates the output predictions.

• The input layer stores the weights of the network.

• The input layer receives and processes the input data for the neural network.

What is overfitting in the context of neural networks?

• Overfitting is when a neural network performs well on training data but poorly on unseen data due to excessive learning of noise.

• Overfitting is when a model generalizes well to both training and unseen data.

• Overfitting happens when a neural network is undertrained and lacks complexity.

• Overfitting occurs when a neural network has too few parameters.

Name one common activation function used in neural networks.

• ReLU

• Softmax

• Sigmoid

• Tanh

What does the output layer of a neural network do?

• The output layer processes input data for training.

• The output layer adjusts weights during backpropagation.

• The output layer stores the model parameters.

• The output layer generates the final predictions of the neural network.

How can you improve the performance of a neural network?

• Ignore validation data during training

• Use random noise as input data

• Use data augmentation, regularization, optimize architecture, tune hyperparameters, and use better optimization algorithms.

• Increase the number of layers without purpose

What is the difference between supervised and unsupervised learning?

• Supervised learning requires no data for training, while unsupervised learning requires all data to be labeled.

• Supervised learning is faster than unsupervised learning because it uses less data.

• Unsupervised learning is only used for classification tasks, while supervised learning is used for regression tasks.

• Supervised learning uses labeled data for training, while unsupervised learning uses unlabeled data to find patterns.