Open Access Week is on and this year’s motto is “Open in Action”. Thus we’ll take the chance to feature a really interesting open science project we’ve recently stumbled upon: OpenML. OpenML sets out (and actually already achieved) to make machine learning available to a broader audience (especially scientists) and build a platform to create, share, evaluate and use machine learning algorithms. We took the chance to talk to Heidi Seibold and Joaquin Vanschoren about the project’s history, current state and future plans! Feel invited to give OpenML a chance, test it, or contribute to it. For now, enjoy this episode!
P.S.: we had to partly switch to a backup recording option, hopefully you don’t mind too much.
Episode Info: Duration 0:48:23 Recorded on 21-10-2016, Published on 28-10-2016
This work is licensed under a Creative Commons Attribution 4.0 International License
- Download:
- MP3 Audio33 MB
- MP4 Audio35 MB
- OPUS Audio36 MB
The Basics of OpenML
00:06:36Polymath; — Tim Gowers; — Bernd Bischl; — R Project for Statistical Computing; — OpenML Github Repository; — OpenML Community on Google groups;.
Working with and on OpenML
00:18:34Contributions; — Github DOI; — Altmetric; — Users statistics; — Comparing and evaluating machine learning algorithms; — International Conference on Machine Learning ICML; — European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases ECMLPKDD;.
OpenML's Future
00:40:12Linked Open Data; — Integrate OpenML into machine learning tools; — Funding Opportunities;.