A stacked autoencoder (SAE) [16,17] stacks multiple AEs to form a deep structure. It feeds the hidden layer of the kth AE as the input feature to the ...
確定！ 回上一頁
