Probabilistic Linear Solvers for Machine Learning
From MaRDI portal
Publication:6351692
arXiv2010.09691MaRDI QIDQ6351692
Author name not available (Why is that?)
Publication date: 19 October 2020
Abstract: Linear systems are the bedrock of virtually all numerical computation. Machine learning poses specific challenges for the solution of such systems due to their scale, characteristic structure, stochasticity and the central role of uncertainty in the field. Unifying earlier work we propose a class of probabilistic linear solvers which jointly infer the matrix, its inverse and the solution from matrix-vector product observations. This class emerges from a fundamental set of desiderata which constrains the space of possible algorithms and recovers the method of conjugate gradients under certain conditions. We demonstrate how to incorporate prior spectral information in order to calibrate uncertainty and experimentally showcase the potential of such solvers for machine learning.
Has companion code repository: https://github.com/JonathanWenger/probabilistic-linear-solvers-for-ml
This page was built for publication: Probabilistic Linear Solvers for Machine Learning
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6351692)
