Data Association with Gaussian Processes

TL;DR

This paper introduces a Bayesian method using Gaussian processes to solve data association problems, effectively handling noisy, multimodal data sources while enabling simultaneous supervised learning.

Contribution

It proposes a novel fully Bayesian approach with Gaussian process priors for data association, including an efficient inference scheme adaptable to deep Gaussian processes.

Findings

Effective data association in noisy, multimodal data environments

Unified framework for data association and supervised learning

Scalable inference method using doubly stochastic variational inference

Abstract

The data association problem is concerned with separating data coming from different generating processes, for example when data come from different data sources, contain significant noise, or exhibit multimodality. We present a fully Bayesian approach to this problem. Our model is capable of simultaneously solving the data association problem and the induced supervised learning problems. Underpinning our approach is the use of Gaussian process priors to encode the structure of both the data and the data associations. We present an efficient learning scheme based on doubly stochastic variational inference and discuss how it can be applied to deep Gaussian process priors.