Question# 01: Consider the simple neural network below: 0.1 Input A = 0.1 0.2 0.5 Output 0.3 Input B= 0.7 0.1 0.2 Assume that the neurons have a Sigmoid activation function, Target value is 1 and learning rate is also 1. Perform the following task: a) Perform a forward pass on the network. b) Perform a reverse pass (training) once. c) Perform a further forward pass and comment on the result.

A machine learning model is learning parameters w₁ and w₂. It turns out that for the latest data inputted, the error function is: E=(w₁-1)²+(W₂-1)² The current values of the parameters are w₁-2 and w₂=0.9. Gradient descent is applied to update the parameters. Which statement is true? 7 0 w1 2 W2 3 4 0 W2 W1 decreases and decreases, change in w₁ is bigger. W₁ decreases and w₂ increases, change in w₂ is bigger. W₁ decreases and w₂ increases, change in w₁ is bigger. W₁ decreases and w₂ increases, both by the same amount. 18 16 14 12 10 8 6 4 HH +++ 2