Learning Target Vi 51 4484 Softmax V2 is 2 :: 150 16x16=256 lnk→1 The learning target is ..... No ink→0 Input: y has the maximum value Input: y>has the maximum value 口卡B·三4色进分双0
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Learning Target
A good function should make the loss Loss of all examples as small as possible. As close as 1 possible Given a set of Y2 0 parameters Softmax ....: L055 yi0 0 Loss can be square error or cross entropy target between the network output and target 4口◆4⊙t1三1=，￥9QC
. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..
Total Loss: Total Loss For all training data… NN As small as possible NN Find a function in function set that NN minimizes total loss L Find the network parameters 0'that NN minimize total loss L 1口卡+B·定，色进分QG
. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..
Table of Contents 深度学习简介 Neural Network Goodness of Function Pick the Best Function 前馈神经网络 Tips for Deep Learning 卷积神经网络(Convolutional Neural Network,CNN) 循环神经网络(Recurrent Neural Network,RNN Keras CNN in Keras RNN in Keras 4口卡404三·1=生0C
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table of Contents 深度学习简介 Neural Network Goodness of Function Pick the Best Function 前馈神经网络 Tips for Deep Learning 卷积神经网络（Convolutional Neural Network, CNN） 循环神经网络（Recurrent Neural Network, RNN） Keras CNN in Keras RNN in Keras
Three Steps for Deep Learning Step 1:define a set of function Step 2:goodness of function Step 3:pick the best function 口·三4，进分双0
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Three Steps for Deep Learning