> For the complete documentation index, see [llms.txt](https://aalborg-university.gitbook.io/machine-learning-workstation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://aalborg-university.gitbook.io/machine-learning-workstation/useful-links/machine-learning/autoencoders.md).

# Autoencoders

**An Autoencoder (AE)** is a Neural Network that can be trained in an Unsupervised fashion (basically Self Supervised). In an AE the Output values are equal to the Inputs. Autoencoders are used to reduce the size of a set of inputs, by generating a more compact representation. The AE consists of an Encoder that compress the data and a Decoder that reconstruct the data from this compressed representation.

* [The Official Keras tutorial about Autoencoders](https://blog.keras.io/building-autoencoders-in-keras.html)
* [A Gentle introduction to Autoencoders ](https://medium.com/edureka/autoencoders-tutorial-cfdcebdefe37)[Using AE to remove noise](https://medium.com/activating-robotic-minds/how-to-reduce-image-noises-by-autoencoder-65d5e6de543)
* [Convolutional AE and more](https://medium.com/activating-robotic-minds/how-to-reduce-image-noises-by-autoencoder-65d5e6de543)
* [More advanced topics with AE](https://bit.ly/2PrmbAG)
* [Introduction to sparce AE and PCA vs AE comparison](https://www.jeremyjordan.me/autoencoders/)
* [Training VAEs](https://becominghuman.ai/variational-autoencoders-for-new-fruits-with-keras-and-pytorch-6d0cfc4eeabd)

#### Sound related

* [Introduction to Magenta Variational Autoencoder](https://magenta.tensorflow.org/music-vae)