You will use a probe with a frequency of 10 MHz to scan the Liver.

The next three items on the exam are different parts of one question (Parts A, B, and C) and are based on the same group of organisms and traits.

Backpropagation. Indicate whether each of the following statements is True or False (1 pt each). (a) Backpropagation relies on repeated application of the chain rule to compute gradients. [a] (b) In backpropagation, gradients are propagated from the input layer to the output layer. [b] (c) In Pytorch, the purpose of calling optimizer.zero_grad() in a training loop is to retain and accumulate the gradients from the previous backward pass. [c] (d) The softmax function is differentiable and thus compatible with backpropagation. [d]

Longshore currents: a) Under what conditions do longshore currents form?  b) Describe 2 ways in which rip currents are different from longshore currents.      Do not simply give definitions of each one.  Instead, explain how they are different. 1.  2. 

Multi-layer perceptron. Consider the following neural network defined in PyTorch.  class NeuralNetwork(nn.Module):    def __init__(self):        super().__init__()        self.linear_relu_stack = nn.Sequential(            nn.Linear(20, 100),            nn.ReLU(),            nn.Linear(100, 100),            nn.ReLU(),            nn.Linear(100, 3),        )    def forward(self, x):        logits = self.linear_relu_stack(x)        return logits (a) (2 pts) How many learnable layers does the neural network have? Count only layers that contain trainable parameters. [a] (b) (2 pts) How many parameters does the neural network have? You may disregard bias/offset terms. [b]

Overfitting. When training a neural network, which of the following techniques does NOT help reduce overfitting? 

Summer25_MAT239_FinalExam.pdf

What is the output? matrix = [[0, 1], [2, 3], [4, 5]]  for row in matrix:      print(row[1], end=’ ‘)

What is the output? lst = [10, 20, 30]  lst.insert(1, 15)  print(lst)

What is the output? for i in range(5):      if i == 4:          break      print(i)