File Exchange

image thumbnail

Denoising Autoencoder

version 1.5.0 (735 KB) by BERGHOUT Tarek
In this code a full version of denoising autoencoder is presented.


Updated 26 Jun 2019

View License

An autoencoder is a type of artificial neural network used to learn efficient data (codings) in an unsupervised manner. The aim of an auto encoder is to learn a representation (encoding) for a set of data,
denoising autoencoders is typically a type of autoencoders that trained to ignore “noise’’ in corrupted input samples.
In this code we represent to you a denoising autoencoder with a single hidden layer feed forward networks trained by Extreme learning machine.
This algorithm allows training and testing of any dataset with the user defined parameters and shows the main results of both of them.

in this version noise is added randomly by frames(blocks of data) , if you need any information or help concerning your deep learning projects contact me via:

Cite As

BERGHOUT Tarek (2020). Denoising Autoencoder (, MATLAB Central File Exchange. Retrieved .

Comments and Ratings (4)


Thank you its very usefull, please i want to apply this autoencoder for channel equalizer can you help me with that?

Thank you and it is of great use.

Thanks it is quite helpful


thanks a lot for sharing



After completing the training process,we will no longer in need To use old Input Weights for mapping the inputs to the hidden layer, and instead of that we will use the Outputweights beta for both coding and decoding phases and.


some coments are added


a new version that trains an autoencoders by adding random samples of noise in each frame (block of data) .


new version


a new illustration image is description notes Note were added

MATLAB Release Compatibility
Created with R2013b
Compatible with any release
Platform Compatibility
Windows macOS Linux

Inspired by: Autoencoders