Quiz 3 Answers

pdf

School

San Jose State University *

*We aren’t endorsed by this school

Course

258

Subject

Electrical Engineering

Date

Jan 9, 2024

Type

pdf

Pages

5

Uploaded by MagistratePorpoiseMaster286

Report
12/7/18, 4 : 04 PM Quiz #3: FA18: EE-258 Sec 01 - Neural Networks Page 1 of 5 https://sjsu.instructure.com/courses/1267065/quizzes/1280503?headless=1 ! This quiz has been regraded; your new score reflects 3 questions that were affected. Quiz #3 Due Dec 4 at 11:59pm Points 10 Questions 10 Available after Dec 3 at 12pm Time Limit 10 Minutes A ! empt History Attempt Time Score Regraded LATEST Attempt 1 3 minutes 8 out of 10 9 out of 10 Score for this quiz: 9 out of 10 Submitted Dec 4 at 6:04pm This attempt took 3 minutes. 1 / 1 pts Question 1 In L2 regularization, the weights are shrinked compared to the unregularized model. True Correct! Correct! False 1 / 1 pts Question 2 A modified version of normal equations can be used to find the weight vector for a linear regression model under L1 regularization.
12/7/18, 4 : 04 PM Quiz #3: FA18: EE-258 Sec 01 - Neural Networks Page 2 of 5 https://sjsu.instructure.com/courses/1267065/quizzes/1280503?headless=1 True False Correct! Correct! 1 / 1 pts Question 3 Dataset augmentation is a powerful regularization technique for image classification tasks. True Correct! Correct! False 1 / 1 pts Question 4 The weights of the kernel in a convolution layer of CNNs are preset before training of the neural networks. True False Correct! Correct! 1 / 1 pts Question 5 Bagging uses sampling with replacement to generate multiple different datasets.
12/7/18, 4 : 04 PM Quiz #3: FA18: EE-258 Sec 01 - Neural Networks Page 3 of 5 https://sjsu.instructure.com/courses/1267065/quizzes/1280503?headless=1 True Correct! Correct! False 1 / 1 pts Question 6 Dropout trains the ensemble consisting of all subnetworks that can be formed by removing nonoutput units from an underlying base network. True Correct! Correct! False 1 / 1 pts Question 7 ResNet is a convolutional neural network. True Correct! Correct! False 1 / 1 pts Question 8 A CNN has many fewer parameters than a fully connected DNN.
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
12/7/18, 4 : 04 PM Quiz #3: FA18: EE-258 Sec 01 - Neural Networks Page 4 of 5 https://sjsu.instructure.com/courses/1267065/quizzes/1280503?headless=1 True Correct! Correct! False 0 / 1 pts Question 9 Consider a CNN composed of three convolutional layers, each with 3x3 kernels, a stride of 2, and SAME padding. The lowest layer outputs 10 feature maps, the middle one outputs 20, and the top one outputs 40. The input image are gray-scale images of 20x30 pixels. What is the number of parameters per feature map in the first layer? 28 You Answered You Answered 100 280 10 Correct Answer Correct Answer Original Score: 0 / 1 pts Regraded Score: 1 / 1 pts Question 10 ! This question has been regraded. Consider a CNN composed of three convolutional layers, each with 3x3 kernels, a stride of 2, and SAME padding. The lowest layer outputs 10 feature maps, the middle one outputs 20, and the top one outputs 40. The input image are gray-scale images of 20x30 pixels. What is the total number of parameters in the second layer?
12/7/18, 4 : 04 PM Quiz #3: FA18: EE-258 Sec 01 - Neural Networks Page 5 of 5 https://sjsu.instructure.com/courses/1267065/quizzes/1280503?headless=1 90 91 You Answered You Answered 900 910 1820 Correct Answer Correct Answer Quiz Score: 9 out of 10