Heteroscedastic Relevance Vector Machine

TL;DR

This paper introduces a heteroscedastic extension to the Relevance Vector Machine (RVM) for regression, utilizing variational approximation and expectation propagation to handle varying noise levels in data.

Contribution

It presents a novel heteroscedastic RVM model that incorporates advanced Bayesian inference techniques, improving regression modeling for data with non-constant noise.

Findings

Model under development, results pending comparison

Uses variational approximation and expectation propagation

Aims to improve regression with heteroscedastic noise

Abstract

In this work we propose a heteroscedastic generalization to RVM, a fast Bayesian framework for regression, based on some recent similar works. We use variational approximation and expectation propagation to tackle the problem. The work is still under progress and we are examining the results and comparing with the previous works.