Auto-encoder -Pre-training DNN Find-tune by Greedy Layer-wise Pre-training again backpropagation output 10 output 10 Random ↓合 W4 init 500 500 136JDL 合 ↓合w 1000 1000 ↓香w2 1000 1000 合 ↓1w Input 784 Input 784 Auto-encoder – Pre-training DNN • Greedy Layer-wise Pre-training again Target 784 1000 1000 10 Input output Input 784 1000 W1 1000 𝑥 W2 W3 500 500 10 output W4 Random init Find-tune by backpropagation