ReproducedPapers.org: Openly teaching and structuring machine learning reproducibility
Burak Yildiz  1  , Hayley Hung  1  , Jesse H. Krijthe  1  , Cynthia C. S. Liem  1  , Marco Loog  1  , Gosia Migut  1  , Oliehoek Frans  1  , Annibale Panichella  1  , Przemysław Pawełczak  1  , Stjepan Picek  1  , Mathijs De Weerdt  1  , Jan Van Gemert  1  
1 : Department of Electrical Engineering, Mathematics and Computer Science [Delft]  -  Website
Delft University of Technology, Postbus 5 2600 AA Delft -  Netherlands

We present ReproducedPapers.org: an open online repository for teaching and structuring machine learning reproducibility. We evaluate the effect of doing a reproduction project among students and the added value of an online reproduction repository among AI researchers. We use anonymous self-assessment surveys and obtained 144 responses. Results suggest that students who do a reproduction project place more value on scientific reproductions and become more critical thinkers. Students and AI researchers agree that our online reproduction repository is valuable.



  • Poster
Online user: 3 RSS Feed | Privacy
Loading...