Fitting Laplacian Regularized Stratified Gaussian Models

05/04/2020
by   Jonathan Tuck, et al.
1

We consider the problem of jointly estimating multiple related zero-mean Gaussian distributions from data. We propose to jointly estimate these covariance matrices using Laplacian regularized stratified model fitting, which includes loss and regularization terms for each covariance matrix, and also a term that encourages the different covariances matrices to be close. This method `borrows strength' from the neighboring covariances, to improve its estimate. With well chosen hyper-parameters, such models can perform very well, especially in the low data regime. We propose a distributed method that scales to large problems, and illustrate the efficacy of the method with examples in finance, radar signal processing, and weather forecasting.

READ FULL TEXT

page 10

page 11

page 18

research
04/26/2019

A Distributed Method for Fitting Laplacian Regularized Stratified Models

Stratified models are models that depend in an arbitrary way on a set of...
research
05/04/2023

Joint Graph Learning and Model Fitting in Laplacian Regularized Stratified Models

Laplacian regularized stratified models (LRSM) are models that utilize t...
research
01/27/2020

Eigen-Stratified Models

Stratified models depend in an arbitrary way on a selected categorical f...
research
11/07/2011

Discriminant Analysis with Adaptively Pooled Covariance

Linear and Quadratic Discriminant analysis (LDA/QDA) are common tools fo...
research
03/27/2023

Regularized EM algorithm

Expectation-Maximization (EM) algorithm is a widely used iterative algor...
research
05/30/2020

Scale matrix estimation under data-based loss in high and low dimensions

We consider the problem of estimating the scale matrix Σ of the additif ...
research
12/10/2022

Spatiotemporal Residual Regularization with Dynamic Mixtures for Traffic Forecasting

Existing deep learning-based traffic forecasting models are mainly train...

Please sign up or login with your details

Forgot password? Click here to reset