Determinantal point processes (DPPs) were introduced by Macchi (Macchi 1975
, 83-122) as a model for repulsive (fermionic) particle distributions. But their recent popularization is largely due to their usefulness for encouraging diversity in the final stage of a recommender system (Kulesza & Taskar 2012
, 123-286). The standard sampling scheme for finite DPPs is a spectral decomposition followed by an equivalent of a randomly diagonally pivoted Cholesky factorization of an orthogonal projection, which is only applicable to Hermitian kernels and has an expensive set-up cost. Researchers Launay
2018 (http://arxiv.org/abs/1802.08429); Chen & Zhang 2018
(https://papers.nips.cc/paper/7805-fast-greedy-map-inference-for-determinantal-point-process-to-improve-recommendation-diversity.pdf) have begun to connect DPP...

High-performance sampling of generic determinantal point processes

10.1098/rsta.2019.0059