Suppose you have a 2-layer neural network (1 hidden and 1 output). For the hidden layer,...

 

Transcribed Image Text:

Suppose you have a 2-layer neural network (1 hidden and 1 output). For the hidden layer, the weight matrix is W[1] = [3,4 4.8 and the bias 3.4 6.4 is 6[1] For the output layer, the weight matrix is W2] = [−2.4 0.6], and the bias is 6[2] = [1] 3.4 6.41 rows of W, .., W = [3,4 1 4.8 If we swap the two 8] . and swap the two columns of W2], i.e., W2] = [0.6 -2.4]. How will this change the output of the network, y = g(W2g(W[¹¹x+b[¹])+b²])? Suppose you have a 2-layer neural network (1 hidden and 1 output). For the hidden layer, the weight matrix is W[1] = [3,4 4.8 and the bias 3.4 6.4 is 6[1] For the output layer, the weight matrix is W2] = [−2.4 0.6], and the bias is 6[2] = [1] 3.4 6.41 rows of W, .., W = [3,4 1 4.8 If we swap the two 8] . and swap the two columns of W2], i.e., W2] = [0.6 -2.4]. How will this change the output of the network, y = g(W2g(W[¹¹x+b[¹])+b²])?