Scribd
Upload
1K views3 pages

Machine Learning Andrew NG Week 5 Quiz 1

This document discusses neural networks and backpropagation. It contains a quiz with questions about vectorizing the update step in backpropagation, recovering weight matrices from a flattened vector, numerically computing a derivative using a formula, and statements about gradient checking.

Uploaded by

Hương Đặng
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
1K views3 pages

Machine Learning Andrew NG Week 5 Quiz 1

This document discusses neural networks and backpropagation. It contains a quiz with questions about vectorizing the update step in backpropagation, recovering weight matrices from a flattened vector, numerically computing a derivative using a formula, and statements about gradient checking.

Uploaded by

Hương Đặng
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Coursera: Machine Learning (Week

5) Quiz - Neural Networks: Learning


| Andrew NG

1. You are training a three layer neural network and would like to use backpropagation to
compute the gradient of the cost function. In the backpropagation algorithm, one of the steps is to
update
for every i,j. Which of the following is a correct vectorization of this step?

o  

o  

o  

o  
 This version is correct, as it takes the “outer product” of the two vectors   and   
which is a matrix such that the (i,j)-th entry is as   desired.

2. Suppose Theta1 is a 5x3 matrix, and Theta2 is a 4x6 matrix. You set thetaVec = [Theta1(:),
Theta2(:)]. Which of the following correctly recovers ?
o  reshape(thetaVec(16 : 39), 4, 6)
 This choice is correct, since Theta1 has 15 elements, so Theta2 begins at
index 16 and ends at index 16 + 24 - 1 = 39.
3.

o  reshape(thetaVec(15 : 38), 4, 6)

o  reshape(thetaVec(16 : 24), 4, 6)

o  reshape(thetaVec(15 : 39), 4, 6)

o  reshape(thetaVec(16 : 39), 6, 4)

3. Let  . Let  , and  . Use the formula   to numerically compute


an approximation to the derivative at  . What value do you get? (When  , the true/exact
derivati ve is  .)

o  8

o  6.0002

 We compute  .
o  6

o  5.9998

4. Which of the following statements are true? Check all that apply.

o  For computational efficiency, after we have performed gradient checking to


verify that our backpropagation code is correct, we usually disable gradient checking before
using backpropagation to train the network.
 Checking the gradient numerically is a debugging tool: it helps ensure a correct
implementation, but it is too slow to use as a method for actually computing gradients.
5.

o  Computing the gradient of the cost function in a neural network has the same
efficiency when we use backpropagation or when we numerically compute it using the
method of gradient checking.

o  Using gradient checking can help verify if one’s implementation of


backpropagation is bug-free.
 If the gradient computed by backpropagation is the same as one computed numerically with
gradient checking, this is very strong evidence that you have a correct implementation of
backpropagation.

o  Gradient checking is useful if we are using one of the advanced optimization


methods (such as in fminunc) as our optimization algorithm. However, it serves little purpose
if we are using gradient descent.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy