Hi,

There's 3 or more maybe, simple Autoencoder, denoising Autoencoder, variational Autoencoder. When they're stacked we have a stacked Autoencoder.

Respected Dairi Abdelkader/ Luaay Abdulwahid Shihab ,

Thank you so much for your answer...

I found one more type of Auto-encoder i.e. Sparse Auto-encoder..

You're welcome as told you there are 3 or more.. They are extending by adding or noise sparse or adding latentes variables or changing the learning approach

https://codeburst.io/deep-learning-types-and-autoencoders-a40ee6754663

https://towardsdatascience.com/deep-inside-autoencoders-7e41f319999f

https://www.erastrat.com/tag/types-of-autoencoders/

Respected Mohamed AbdElAziz Khamis,

Thank you so your answer...

We recently developed a new type of auto-encoder based on the Wasserstein distance (paper attached). It's called the Sliced-Wasserstein Auto-Encoder and it also acts as a generative model.

Respected Charles E. Martin,

Thank you so much for your reference..