a 2. Neural Networks. Consider a neural network, with No = 2 input units, x = (11, 12), Nn = 3 hidden units with ReLU ac

[answerhappygod]

A 2 Neural Networks Consider A Neural Network With No 2 Input Units X 11 12 Nn 3 Hidden Units With Relu Ac 1 (44.39 KiB) Viewed 40 times

a 2. Neural Networks. Consider a neural network, with No = 2 input units, x = (11, 12), Nn = 3 hidden units with ReLU activations and one output unit for regression, 2= wkrk + b), u) = max{0,2"}, ;= 1,...,Nh N И k=1 N VO v=Ewệu + bº, k=1 (a) Suppose that w" = 1 0 0 1 -1 b" -0.5 -1 3 Write equations for 2y in terms of r for j = 1,2,3. (b) Draw the region of inputs (11, 12) where us > 0 for all j. (c) You are given data x, y as well as weights and biases wh, bh for the hidden layer. Write a few lines of python code to fit wo, bo for the output layer by minimizing the MSE. You may assume you have a function beta = lstsq(,b) # Solves the least squares problem A.dot (beta) = b