Cisco®
CompTIA
Checkpoint
ISC
EC-Council
EXIN
Fortinet
ITIL®
Juniper
Microsoft
VMware
Avaya
SAP
Google
Salesforce
Amazon
Palo Alto Networks
ISC2
All Vendors
  2. Home
  3. Exams
  4. CompTIA
  5. DY0-001

Free CompTIA DY0-001 Exam Questions (page: 9)

Page 3 of 18
PDF with 85 Questions

Which of the following explains back propagation?

  1. The passage of convolutions backward through a neural network to update weights and biases
  2. The passage of accuracy backward through a neural network to update weights and biases
  3. The passage of nodes backward through a neural network to update weights and biases
  4. The passage of errors backward through a neural network to update weights and biases

Answer(s): D

Explanation:

Back propagation computes the gradient of the loss (error) with respect to each weight by propagating the error signal backward through the network, then uses those gradients to adjust weights and biases.



A data scientist is building a proof of concept for a commercialized machine-learning model.
Which of the following is the best starting point?

  1. Literature review
  2. Model performance evaluation
  3. Hyperparameter tuning
  4. Model selection

Answer(s): A

Explanation:

Before diving into selecting or tuning models, a literature review grounds the proof of concept in existing research and best practices, ensuring the approach aligns with state-of-the-art methods and the problem's domain requirements.



Which of the following best describes the minimization of the residual term in a LASSO linear regression?

  1. |e|
  2. e
  3. 0
  4. e2

Answer(s): D

Explanation:

LASSO regression retains the ordinary least squares loss by minimizing the sum of squared residuals (e²), with an added L1 penalty on the coefficients, but the residual term itself remains squared.



Which of the following layer sets includes the minimum three layers required to constitute an artificial neural network?

  1. An input layer, a pooling layer, and an output layer
  2. An input layer, a convolutional layer, and a hidden layer
  3. An input layer, a hidden layer, and an output layer
  4. An input layer, a dropout layer, and a hidden layer

Answer(s): C

Explanation:

By definition, an artificial neural network requires at least these three fundamental layers: the input layer to receive data, one or more hidden layers to perform transformations, and the output layer to produce predictions. Pooling, convolutional, and dropout layers are useful in specialized architectures (e.g., CNNs) but aren't part of the minimal ANN structure.



A data scientist uses a large data set to build multiple linear regression models to predict the likely market value of a real estate property. The selected new model has an RMSE of 995 on the holdout set and an adjusted R2 of .75. The benchmark model has an RMSE of 1,000 on the holdout set.
Which of the following is the best business statement regarding the new model?

  1. The model should be deployed because it has a lower RMSE.
  2. The model's adjusted R2 is exceptionally strong for such a complex relationship.
  3. The model fails to improve meaningfully on the benchmark model.
  4. The model's adjusted R2 is too low for the real estate industry.

Answer(s): C

Explanation:

Although the new model's RMSE is technically lower (995 vs. 1,000), the five-point improvement on holdout data is negligible in most real-estate contexts and unlikely to produce meaningful business value over the existing benchmark.



Viewing page 9 of 18



Post your Comments and Discuss CompTIA DY0-001 exam prep with other Community members:

DY0-001 Exam Discussions & Posts