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.
@article{arxiv.2008.06716,
title = {Performance of Hyperbolic Geometry Models on Top-N Recommendation Tasks},
author = {Leyla Mirvakhabova and Evgeny Frolov and Valentin Khrulkov and Ivan Oseledets and Alexander Tuzhilin},
journal= {arXiv preprint arXiv:2008.06716},
year = {2020}
}