VBALD - Variational Bayesian Approximation of Log Determinants

Abstract

Evaluating the log determinant of a positive definite matrix is ubiquitous in machine learning. Applications thereof range from Gaussian processes, minimum-volume ellipsoids, metric learning, kernel learning, Bayesian neural networks, Determinental Point Processes, Markov random fields to partition functions of discrete graphical models. In order to avoid the canonical, yet prohibitive, Cholesky $\mathcal{O}(n^3)$ computational cost, we propose a novel approach, with complexity $\mathcal{O}(n^2)$, based on a constrained variational Bayes algorithm. We compare our method to Taylor, Chebyshev and Lanczos approaches and show state of the art performance on both synthetic and real-world datasets.

References in corpus (7)

• Determinantal point processes for machine learning
• Maximum entropy and the problem of moments: A stable algorithm
• Fast DPP Sampling for Nyström with Application to Kernel Methods
• Scalable Log Determinants for Gaussian Process Kernel Learning
• Improved stochastic trace estimation using mutually unbiased bases
• Bayesian Inference of Log Determinants
• Faster Greedy MAP Inference for Determinantal Point Processes

Cited by in corpus (2)

• A novel approach to visibility-space modelling of interferometric gravitational lens observations at high angular resolution
• Deep Curvature Suite