What is Unsupervised Papers?

Unsupervised Papers is an open-source directory of unsupervised machine learning papers, tasks and methods with links to articles, websites, videos and code.

Who is Unsupervised Papers for?

It is intended to serve a few audiences in particular. First the machine learning researcher or practitioner who would like a quick reference to the state of the art in unsupervised methods will find this site useful. Also, this will be an invaluable tool for students and educators to assist with in-depth research into the field. And lastly, it will be a great resource for hackers and ameture artificial intelligence enthusiasts to dip their toe into what promises to be a field with a very exciting future.

How can I contribute?

The entire directory is free and open-source. This is a collaborative project and contributions are welcome via our GitHub page https://github.com/virtualgraham/unsupervisedpapers. All the entries from the site are encoded as markdown and image files in the repository. This enables the free and open, but moderated system of changes that Git was built for.

Criteria for Inclusion

Generally papers included in the Unsupervised Papers index meet the following criteria:

• Are primally focused on unsupervised or self-supervised learning methods or cited as components to these methods.
• Has a freely available PDF file of the full text
• Is in the Semantic Scholar corpus

Have questions or feedback? Contact me at virtualgraham@unsupervisedpapers.com.