CECS 456 Midterm 2 Exam Questions And Answers 100% Correct Verified Guaranteed Answers.

For the following code: model = Sequential([ Dense(units=25, activations="sigmoid"), Dense(units=15, activations="sigmoid"), Dense(units=10, activations="sigmoid"), Dense(units=1, activations="sigmoid")]) This code will define a neural network with how many layers? A) 3 B) 5 C) 4 D) 25 - correct answer C) 4 Which of the following is true: A) w^[4]_3 is the column vector of parameters of the third layer and fourth neuron B) a^[2]_3 denotes the activation vector of the second layer for the third example C) a^[2] denotes the activation vector of the second layer D) w^[4]_3 is the row vector of parameters of the fourth layer and third neuron - correct answer C) a^[2] denotes the activation vector of the second layer Image 1: What are the dimensions of Z^[1] and A^[1]? A) (4,1 ) B) (4, m) C) (4,2) D) (1,4) - correct answer C) (4,2) The use of the ReLU activation function is becoming more rare because the ReLU function has no derivative for c = 0. A) True B) False - correct answer B) False In which of the following cases in the linear (identity) activation function most likely used? A) for binary classification problems B) the linear activation function is never used C) as activation function in the hidden layers D) when working with regression problems - correct answer D) when working with regression problems Image 2: For the multiclass classification, the recommended way to implement softmax regression is to set from_logits=True in the loss function, and also to define the model's output layer with... A) a 'linear' activation B) a 'softmax' activation - correct answer A) a 'linear' activation Image 3: For which type of task would you use the binary cross entropy loss function? A) BinaryCrossentropy() should not be used for any task B) binary classification (classification with exactly 2 classes) C) regression tasks (tasks that predict a number) D) a classification task that has 3 or more classes (categories) - correct answer B) binary classification (classification with exactly 2 classes) Image 4: For the task of predicting housing prices, which activation functions could you choose for the output layer? Choose 2 options that apply. A) linear B) sigmoid C) ReLU - correct answer A) linear C) ReLU Which of the following activation functions is the most common choice for the hidden layers of a neural network? A) most hidden layers do not use any activation function B) sigmoid C) ReLU (rectified linear unit) D) Linear - correct answer C) ReLU (rectified linear unit) In the context of machine learning, what is a diagnostic? A) this refers to the process of measuring how well a learning algorithm does on a test set (data that the algorithm was not trained on). B) a test that you run to gain insight into what is/isn't working with a learning algorithm C) a process by which we quickly try to as many different ways to improve an algorithm as possible, so as to see what works D) an application of machine learning to medical applications, with the goal of diagnosing patients' conditions - correct answer B) a test that you run to gain insight into what is/isn't working with a learning algorithm It is always true that the better an algorithm does on the training set, the better it will do on generalizing to new data