### TensorflowDeep LearningMathematical Optimizationhyperparameter tuningAlso Check: How to Apply for Coursera Financial Aid Course Link: https://www.coursera.org/learn/deep-neural-network

Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization Week 1 Quiz Answers - Coursera!

Practice Exercise - Practical aspects of Deep Learning

Question 1) If you have 10,000,000 examples, how would you split the train/dev/test set?

•  98% train. 1% dev. 1% test
•  60% train . 20% dev . 20% test
•  33% train . 33% dev . 33% test

Question 2) The dev and test set should:
•
• Come from the same distribution
•  Have the same number of examples
•  Come from different distributions
•  Be identical to each other (same (x,y) pairs)

Question 3) If your Neural Network model seems to have high bias, what of the following would be promising things to try? (Check all that apply.)

•  Get more test data
•  Get more training data
•  Make the Neural Network deeper
•  Increase the number of units in each hidden layer

Question 4) You are working on an automated check-out kiosk for a supermarket, and are building a classifier for apples, bananas and oranges.

Suppose your classifier obtains a training set error of 0.5%, and a dev set error of 7%. Which of the following are promising things to try to improve your classifier? (Check all that apply.)

•  Get more training data
•  Use a bigger neural network
•  Increase the regularization parameter lambda
•  Decrease the regularization parameter lambda

Question 5) What is weight decay?

•  The process of gradually decreasing the learning rate during training.
•  A technique to avoid vanishing gradient by imposing a ceiling on the values of the weights.
•  Gradual corruption of the dee in the neural network if it is trained on noisy data.
•  A regularization technique (such as L2 regularization) that results in gradient descent shrinking the weights on every iteration.

Question 6) What happens when you increase the regularization hyperparameter lambda?
•  Weights are pushed toward becoming smaller (closer to 0)
•  Doubling lambda should roughly result in doubling the weights
•  Weights are pushed toward becoming bigger (further from 0)
•  Gradient descent taking bigger steps with each iteration (proportional to lambda)

Question 7) With the inverted dropout technique, at test time:
•  You apply dropout (randomly eliminating units) and do not keep the 1/keep_prob factor in the calculations used in training
•  You do not apply dropout (do not randomly eliminate units), but keep the 1/keep_prob factor in the calculations used in training.
•  You do not apply dropout (do not randomly eliminate units) and do not keep the 1/keep_prob factor in the calculations used in training
•  You apply dropout (randomly eliminating units) but keep the 1/keep_prob factor in the calculations used in training

Question 8) Increasing the parameter keep_prob from (say) 0.5 to 0.6 will likely cause the following: (Check the two that apply)

•  Increasing the regularization effect
•  Reducing the regularization effect
•  Causing the neural network to end up with a higher training set error
•  Causing the neural network to end up with a lower training set error

Question 9) Which of these techniques are useful for reducing variance (reducing overfitting)? (Check all that apply.)

•  Xavier initialization
•  Dropout
•  Data augmentation
•  L2 regularization

Question 10) Why do we normalize the inputs x?
•  It makes the parameter initialization faster
•  It makes it easier to visualize the data
•  It makes the cost function faster to optimize
•  Normalization is another word for regularization--It helps to reduce variance

Improving Deep Neural Networks: Hyperparameter Tuning Regularization and Optimization Week 2 Quiz Answers

Practice Exercise - Optimization Algorithms

Question 1) Which notation would you use to denote the 3rd layer’s activations when the input is the 7th example from the 8th minibatch?
• a{7}(8)
• a{7}(3)
• a{8}(7)
• a{3}(7)

Question 2) Which of these statements about mini-batch gradient descent do you agree with?

• Training one epoch (one pass through the training set) using mini-batch gradient descent is faster than training one epoch using batch gradient descent.
• You should implement mini-batch gradient descent without an explicit for-loop over different mini-batches, so that the algorithm processes all mini-batches at the same time (vectorization).
• One iteration of mini-batch gradient descent (computing on a single mini-batch) is faster than one iteration of batch gradient descent.

Question 3) Why is the best mini-batch size usually not 1 and not m, but instead something in-between?
•  If the mini-batch size is m, you end up with stochastic gradient descent, which is usually slower than mini-batch gradient descent.
•  If the mini-batch size is 1, you end up having to process the entire training set before making any progress.
•  If the mini-batch size is 1, you lose the benefits of vectorization across examples in the mini-batch.
•  If the mini-batch size is m, you end up with batch gradient descent, which has to process the whole training set before making progress. ---> Correct
•  Suppose your learning algorithm’s cost J, plotted as a function of the number of iterations, looks like this:

Question 4) Which of the following do you agree with?

•  If you’re using mini-batch gradient descent, something is wrong. But if you’re using batch gradient descent, this looks acceptable.
•  If you’re using mini-batch gradient descent, this looks acceptable. But if you’re using batch gradient descent, something is wrong.
•  Whether you’re using batch gradient descent or mini-batch gradient descent, something is wrong.
•  Whether you’re using batch gradient descent or mini-batch gradient descent, this looks acceptable.

Question 5) Suppose the temperature in Casablanca over the first three days of January are the same:

Jan 1st: θ1=10oC
Jan 2nd: θ210oC

(We used Fahrenheit in lecture, so will use Celsius here in honor of the metric world.)

Say you use an exponentially weighted average with β=0.5 to track the temperature: v0=0, vt=βvt−1+(1−β)θt. If v2 is the value computed after day 2 without bias correction, and v2corrected is the value you compute with bias correction. What are these values? (You might be able to do this without a calculator, but you don't actually need one. Remember what is bias correction doing.)

• v2=7.5, v2corrected=7.5
• v2=10, v2corrected=7.5
• v2=7.5, v2corrected=10
• v2=10, v2corrected=10

Question 6) Which of these is NOT a good learning rate decay scheme? Here, t is the epoch number.
• α=(t^-.5)α0
• α=(1/1+2*t).α0
• α=(0.95^t).α0
• α=(e^t).α0

Question 7) You use an exponentially weighted average on the London temperature dataset. You use the following to track the temperature: vt=βvt−1+(1−β)θt. The red line below was computed using β=0.9. What would happen to your red curve as you vary β? (Check the two that apply)
Decreasing β will shift the red line slightly to the right.

•  Increasing β will create more oscillations within the red line.
•  Increasing β will shift the red line slightly to the right.
•  Decreasing β will create more oscillation within the red line.

Question 8) Consider this figure:

These plots were generated with gradient descent; with gradient descent with momentum (β = 0.5) and gradient descent with momentum (β = 0.9). Which curve corresponds to which algorithm?

• (1) is gradient descent with momentum (small β), (2) is gradient descent with momentum (small β), (3) is gradient descent
• (1) is gradient descent. (2) is gradient descent with momentum (large β) . (3) is gradient descent with momentum (small β)
• (1) is gradient descent with momentum (small β). (2) is gradient descent. (3) is gradient descent with momentum (large β)
• (1) is gradient descent. (2) is gradient descent with momentum (small β). (3) is gradient descent with momentum (large β)

Question 9) Suppose batch gradient descent in a deep network is taking excessively long to find a value of the parameters that achieves a small value for the cost function J(W,b,...,W[L],b[L]). Which of the following techniques could help find parameter values that attain a small value forJ? (Check all that apply)

• Try better random initialization for the weights
• Try tuning the learning rate α
• Try initializing all the weights to zero

• Adam should be used with batch gradient computations, not with mini-batches.
• The learning rate hyperparameter α in Adam usually needs to be tuned.
• We usually use “default” values for the hyperparameters β1,β2 and ε in Adam (β1=0.9, β2=0.999, ε=10−8)

Improving Deep Neural Networks: Hyperparameter Tuning Regularization and Optimization Week 3 Quiz Answers

Practice Exercise - Hyperparameter tuning, Batch Normalization, Programming Frameworks

Question 1) If searching among a large number of hyperparameters, you should try values in a grid rather than random values, so that you can carry out the search more systematically and not rely on chance. True or False?
• True
• False

Question 2) Every hyperparameter, if set poorly, can have a huge negative impact on training, and so all hyperparameters are about equally important to tune well. True or False?

• True
• False

Question 3) During hyperparameter search, whether you try to babysit one model (“Panda” strategy) or train a lot of models in parallel (“Caviar”) is largely determined by:
• The amount of computational power you can access
• The number of hyperparameters you have to tune
• Whether you use batch or mini-batch optimization
• The presence of local minima (and saddle points) in your neural network

Question 4) If you think β (hyperparameter for momentum) is between on 0.9 and 0.99, which of the following is the recommended way to sample a value for beta?

r = np.random.rand()
beta = 1 - 10 ** (-r - 1)

Question 5) Finding good hyperparameter values is very time-consuming. So typically you should do it once at the start of the project, and try to find very good hyperparameters so that you don’t ever have to revisit tuning them again. True or false?

• True
• False

Question 6) In batch normalization as presented in the videos, if you apply it on the lth layer of your neural network, what are you normalizing?

• z^[l]

Question 7) In the normalization formula, why do we use epsilon?
• To speed up convergence
• To have a more accurate normalization
• In case μ is too small
• To avoid division by zero

Question 8) Which of the following statements about γ and β in Batch Norm are true? Only correct options listed

• They can be learned using Adam, Gradient descent with momentum, or RMSprop, not just with gradient descent.
• They set the mean and variance of the linear variable z l] of a given layer.
• [The optimal values are γ = σ + ε, and . 2 β = μ
• β and γ are hyperparameters of the algorithm, which we tune via random sampling
• There is one global value of and one global value of for each layer, and applies to all the hidden units in that layer.

Question 9) After training a neural network with Batch Norm, at test time, to evaluate the neural network on a new example you should:
•
Use the most recent mini-batch’s value of μ and σ to perform the needed normalizations.
•  Skip the step where you normalize using μ and σ since a single test example cannot be normalized
•  Perform the needed normalizations, use μ and σ^2 estimated using an exponentially weighted average across mini-batches seen during training.
•  If you implemented Batch Norm on mini-batches of (say) 256 examples, then to evaluate on one test example, duplicate that example 256 times so that you’re working with a mini-batch the same size as during training.

Question 10) Which of these statements about deep learning programming frameworks are true? (Check all that apply)

•  Deep learning programming frameworks require cloud-based machines to run.
•  A programming framework allows you to code up deep learning algorithms with typically fewer lines of code than a lower-level language such as Python.
•  Even if a project is currently open source, good governance of the project helps ensure that the it remains open even in the long term, rather than become closed or modified to benefit only one company.