Syntax

Description

Input Arguments

Output Arguments

Algorithms

If the input to an autoencoder is a vector $$x\in {ℝ}^{D_x}$$, then the encoder maps the vectorxto another vector $$z\in {ℝ}^{D^{\left(1\right)}}$$as follows:

where the superscript (1) indicates the first layer. $$h^{\left(1\right)}:{ℝ}^{D^{\left(1\right)}}\to {ℝ}^{D^{\left(1\right)}}$$is a transfer function for the encoder, $$W^{\left(1\right)}\in {ℝ}^{D^{\left(1\right)}\times D_{{}^x}}$$is a weight matrix, and $$b^{\left(1\right)}\in {ℝ}^{D^{\left(1\right)}}$$is a bias vector. Then, the decoder maps the encoded representationzback into an estimate of the original input vector,x, as follows:

上标(2)代表第二个拉yer. $$h^{\left(2\right)}:{ℝ}^{D_x}\to {ℝ}^{D_x}$$is the transfer function for the decoder, $$W^{\left(1\right)}\in {ℝ}^{D_{{}^x}\times D^{\left(1\right)}}$$is a weight matrix, and $$b^{\left(2\right)}\in {ℝ}^{D_x}$$is a bias vector.

See Also

encode|trainAutoencoder