Reproduce “Alternatives to Shallow Autoencoders for Collaborative Filtering” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Mario Mallea, Angela Nebot, Francisco Mugica
The paper tests whether user-to-user relationships can improve EASE collaborative filtering while retaining a closed-form fit. u-EASE learns user-user links, while u-EASE-i combines user- and item-side relationships. On ML-1M, u-EASE-i reaches nDCG@100 of 43.51, versus 40.95 for EASE. The proposed models also improve ranking on Yelp2018, but not Gowalla, so gains are not universal.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.