A repository for summarized machine learning papers (aka The Sparknotes for Papers). Inspired by arXivTimes, a Japanese version of quick summary of papers. This can be considered as the English translateed mirror of arXivTimes.

UPATE: I am currently trying my best to port all of the contents from the original paper. If you know Japanese and would like to help port the contents over, please let me know.

Contents

• Article Summaries
  • A TL;DR or summary of the papers are maintained in the issues section of this repository.
• Datasets
  • All cooresponding datasets which accompany the paper reside here.
• Tools
  • All tools and resources regarding implementation are listed here.
• [Courses]
  • All MOOC courses, YouTube playlists, university lecture pages and other learning resources are provided here.
• [Books]
  • Reading lists for textbook, non-fiction works, and other literature are contained within here.

How to Contribute

To contribute a paper, please follow the format listed below:

• For papers that you would like to contribute, please create a new "Issue".
• Use the name of the paper as the title of the Issues.
• Use the following guidelines when submitting a new entry (please follow the template strictly!):
  • TL;DR - A quick short sentence summary of the paper.
  • Link to the paper.
  • Author/Institution (Copy this from the original paper).

*The general template to be used can be found here.

• The length of the TL;DR should be enough to fit in a single tweet (~140 characters). The "ideal" TL;DR should capture the essence of the problem being solved, the solution/approach the author has taken and the results. Please try your best to help communicate the essence of the paper!
• If you are making a contribution for a specific paper, please designate yourself within the Assignees of the issue. This will help us to identify who has provided content and accordingly give credit.
• Use the Labels to tag the category of the paper accordingly. (Currently only contributors are only allowed to issue those tags, thus we'll take care of the tagging when submissions have been recieved.)
• Use the comments section as a place to discuss, comment, ask questions or give feedback on the paper.

For proposing any meta-level changes to the sparXiv repository, such as adding more tags, changing the template format, please create a new issue using the proposal tag and provide us with your feedback!