A developer is implementing backpropagation manually in a framework that does not provide automatic differentiation. The network has a single hidden layer with ReLU activation. The loss is mean squared error. Which step is essential for correctly updating the weights?

Use the derivative of the ReLU function only for the output layer.

Normalize the inputs to have zero mean and unit variance before each forward pass.

Randomly initialize the weights and keep them fixed during training.

Compute the gradient of the loss with respect to each weight using the chain rule.

What is the primary purpose of dropout in a neural network?

To increase the capacity of the model by adding more neurons to each layer

To normalize the activations of each layer to have zero mean and unit variance

To reduce overfitting by randomly deactivating neurons during training

To speed up training by reducing the number of parameters in the model

A neural network with a single hidden layer is not learning complex patterns in a dataset. The data scientist suspects the network lacks capacity. Which modification would most directly increase the network's capacity?

Add more neurons to the hidden layer.

Add dropout to the hidden layer.

A data scientist is training a deep neural network for image classification. The model overfits on the training data. The team wants to add a regularization technique that also speeds up training. Which combination of techniques would best achieve both goals?

Dropout and batch normalization

Batch normalization and L2 regularization

CompTIADataX

Domain 3Objective 4

Deep Learning DY0-001 Practice Questions (Page 1)

Part of the Machine learning domain, which accounts for 24% of the DY0-001 exam. CompTIA does not publish an official question count, but from its 165-minute exam (~65–110 total, ~16–26 in this domain), expect 3–5 from this objective — we provide 24 practice questions to prepare you well beyond it. (estimate)

24questions here

5free pages

5concepts

24%of the exam

Questions 1–5

1expert · hard
A developer is implementing backpropagation manually in a framework that does not provide automatic differentiation. The network has a single hidden layer with ReLU activation. The loss is mean squared error. Which step is essential for correctly updating the weights?
Select an answer first
2foundation · easy
What is the primary purpose of dropout in a neural network?
Select an answer first
3application · easy
A neural network with a single hidden layer is not learning complex patterns in a dataset. The data scientist suspects the network lacks capacity. Which modification would most directly increase the network's capacity?
Select an answer first
4application · easy
A developer is implementing a neural network in PyTorch. They want to use automatic differentiation to compute gradients for backpropagation. Which PyTorch feature enables this?
Select an answer first
5application · medium
A data scientist is training a deep neural network for image classification. The model overfits on the training data. The team wants to add a regularization technique that also speeds up training. Which combination of techniques would best achieve both goals?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to DY0-001

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by CompTIA. “DY0-001” is a trademark of its owner, used for identification only.