Detecting Nonproperness of Likelihood Equations
2026-08-03 • Machine Learning
Machine LearningSymbolic Computation
AI summaryⓘ
The authors study a problem in algebraic statistics where they want to understand how many ways a data set can have positive 'critical points' related to likelihood, which are solutions to certain equations. They focus on special sets of data, called nonproperness sets, that cause these solutions to behave strangely, like appearing at infinity. The authors develop a new, more efficient method to find these nonproperness sets and prove it works correctly. Their experiments show this method is faster than previous approaches.
algebraic statistical modellikelihood equationspositive critical pointsreal root classificationdiscriminant varietynonproperness setsolutions at infinityalgebraic systemcomputational algebraroot counting
Authors
Xiaoxian Tang, Bican Xia, Tianqi Zhao
Abstract
Given an algebraic statistical model, a challenging problem is classifying the data according to the number of positive critical points of the likelihood function. The positive critical points are the positive solutions to an algebraic system, say likelihood equations. So, identifying the number of positive critical points is a real root classification problem for the likelihood equations. A discriminant variety of a likelihood-equation system geometrically describes the data for which the number of real solutions becomes unusual. As an essential component of the discriminant variety, the nonproperness set collects the data such that the likelihood-equation system has a solution at infinity. So, the number of real solutions varies when the data passes the nonproperness set, and identifying the nonproperness set plays a crucial role in the real root classification. In this work, we develop a novel method for computing nonproperness sets of likelihood-equation systems. We prove the correctness of this method. We show experimentally that it is far more efficient than the known methods in the literature.