Performance of Hyperbolic Geometry Models on Top-N Recommendation Tasks

August 15, 2020 Β· Declared Dead Β· πŸ› ACM Conference on Recommender Systems

πŸ‘» CAUSE OF DEATH: Ghosted
No code link whatsoever

"No code URL or promise found in abstract"

Evidence collected by the PWNC Scanner

Authors Leyla Mirvakhabova, Evgeny Frolov, Valentin Khrulkov, Ivan Oseledets, Alexander Tuzhilin arXiv ID 2008.06716 Category cs.IR: Information Retrieval Cross-listed cs.LG, stat.ML Citations 38 Venue ACM Conference on Recommender Systems Last Checked 4 months ago
Abstract
We introduce a simple autoencoder based on hyperbolic geometry for solving standard collaborative filtering problem. In contrast to many modern deep learning techniques, we build our solution using only a single hidden layer. Remarkably, even with such a minimalistic approach, we not only outperform the Euclidean counterpart but also achieve a competitive performance with respect to the current state-of-the-art. We additionally explore the effects of space curvature on the quality of hyperbolic models and propose an efficient data-driven method for estimating its optimal value.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

πŸ“œ Similar Papers

In the same crypt β€” Information Retrieval

Died the same way β€” πŸ‘» Ghosted