97
27

Learning Signed Determinantal Point Processes through the Principal Minor Assignment Problem

Abstract

Symmetric determinantal point processes (DPP's) are a class of probabilistic models that encode the random selection of items that exhibit a repulsive behavior. They have attracted a lot of attention in machine learning, when returning diverse sets of items is sought for. Sampling and learning these symmetric DPP's is pretty well understood. In this work, we consider a new class of DPP's, which we call signed DPP's, where we break the symmetry and allow attractive behaviors. We set the ground for learning signed DPP's through a method of moments, by solving the so called principal assignment problem for a class of matrices KK that satisfy Ki,j=±Kj,iK_{i,j}=\pm K_{j,i}, iji\neq j, in polynomial time.

View on arXiv
Comments on this paper