GMLE 무료 덤프문제 온라인 액세스

You are building a deep neural network model for image classification, but after training for several epochs, you notice the model is overfitting, with significantly better performance on the training data than the validation data.
What strategies should you apply to reduce overfitting and improve the model's generalization?
Response:

• A.Increase the number of hidden layers and train the model for more epochs
• B.Apply dropout, reduce the number of neurons in each layer, and consider early stopping to prevent overfitting
• C.Reduce the batch size and learning rate
• D.Disable regularization techniques and focus on increasing the dataset size

Which of the following is a supervised learning algorithm?
Response:

• A.k-means clustering
• B.Support Vector Machine (SVM)
• C.Principal Component Analysis (PCA)
• D.Autoencoders

Which is not a typical use case for CNNs?
Response:

• A.Object detection
• B.Image classification
• C.Facial recognition
• D.Natural language processing

What is the main purpose of the 'softmax' function in neural networks?
Response:

• A.To normalize the input data
• B.To speed up the computation in neural networks
• C.To convert the output of the network into probability distributions
• D.To introduce non-linearity into the network

What role does 'dropout' play in training deep neural networks?
Response:

• A.It accelerates the training process
• B.It introduces non-linearity into the model
• C.It helps prevent overfitting by randomly dropping units during training
• D.It increases the accuracy of the model on the training data