(a) do desce -20219 optimizers plots the loss in figure 2 and 3. 20233 0-20242 7130-2021 sc + descent is used for training. Which 2021sc0419 Figure 1 3/07/30-2021sc04196-84446-2023/07/30-0 446-2023 2023/07/30/20216-84440-202 30 Figure 2196-84446-202 /30-2021sc04196-84446-2 04198-84446 (b) In figure 2, assume that the learning rate used was 0.5. 84448-20 Figure 1 effect of increasing the learning rate to 2 and decreasing the learning rate to 0.01. [2] (c) You are given a simple neural network with a single hidden layer containing two neu- rons, and an output layer containing one neuron. All neurons use the sigmoid activation function. The weights and biases of the network are as follows: Redraw the plot to show the 15 to 2 7-20215604196-84/346-20 Weights from input to hidden layer: w = 0.5, W = -0.6 . Biases in hidden layer: bl = 0.1, b2 = -0.2 Weights from hidden layer to output: w3 = 0.7, w4 = -0.8 2. Bias in output layer: b3 = 0.3 20-202 . $0.5 1sc0 130-2 3/07/30 Given an input x = 0.75, calculate the output of the network and mean squared loss if the desired output is 1.25. Use the sigmoid activation function. What will be the effect in the loss if the loss function used is 202 = (218c04196 Z(w,b) }(d y) + N||w2|| 004196-84446- 07/30-2021sc04196-34446 96-844