Latent Bayesian melding for integrating individual and population models

Zhong, Mingjun and Goddard, Nigel and Sutton, Charles (2015) Latent Bayesian melding for integrating individual and population models. In: Advances in Neural Information Processing Systems, 7 - 12 December 2015, Montreal, Canada.

Documents
Latent Bayesian melding for integrating individual and population models
[img]
[Download]
[img]
Preview
PDF
5756-latent-bayesian-melding-for-integrating-individual-and-population-models.pdf - Whole Document

191kB
Item Type:Conference or Workshop contribution (Presentation)
Item Status:Live Archive

Abstract

In many statistical problems, a more coarse-grained model may be suitable for population-level behaviour, whereas a more detailed model is appropriate for accurate modelling of individual behaviour. This raises the question of how to integrate both types of models. Methods such as posterior regularization follow the idea of generalized moment matching, in that they allow matching expectations between two models, but sometimes both models are most conveniently expressed as latent variable models. We propose latent Bayesian melding, which is motivated by averaging the distributions over populations statistics of both the individual-level and the population-level models under a logarithmic opinion pool framework. ln a case study on electricity disaggregation, which is a type of single channel blind source separation problem, we show that latent Bayesian melding leads to significantly more accurate predictions than an approach based solely on generalized moment matching.

Keywords:Bayesian melding, nonintrusive load monitoring
Subjects:G Mathematical and Computer Sciences > G760 Machine Learning
G Mathematical and Computer Sciences > G340 Statistical Modelling
Divisions:College of Science > School of Computer Science
ID Code:27028
Deposited On:24 Apr 2017 09:45

Repository Staff Only: item control page