Bayesian Field Theory: Nonparametric Approaches to Density Estimation, Regression, Classification, and Inverse Quantum Problems

TL;DR

This paper introduces a nonparametric Bayesian framework called Bayesian field theory for learning functions from data, applicable to density estimation, regression, classification, and inverse quantum problems, emphasizing flexibility and numerical solutions.

Contribution

It develops a comprehensive Bayesian field theory approach combining likelihood and prior models, including Gaussian processes, for various complex tasks with a focus on non-Gaussian, computationally feasible solutions.

Findings

Framework applicable to density estimation, regression, classification, and quantum problems.

Inclusion of hyperparameters for flexible prior models.

Growing class of computationally feasible non-Gaussian models.

Abstract

Bayesian field theory denotes a nonparametric Bayesian approach for learning functions from observational data. Based on the principles of Bayesian statistics, a particular Bayesian field theory is defined by combining two models: a likelihood model, providing a probabilistic description of the measurement process, and a prior model, providing the information necessary to generalize from training to non-training data. The particular likelihood models discussed in the paper are those of general density estimation, Gaussian regression, clustering, classification, and models specific for inverse quantum problems. Besides problem typical hard constraints, like normalization and positivity for probabilities, prior models have to implement all the specific, and often vague, "a priori" knowledge available for a specific task. Nonparametric prior models discussed in the paper are Gaussian processes, mixtures of Gaussian processes, and non-quadratic potentials. Prior models are made flexible by including hyperparameters. In particular, the adaption of mean functions and covariance operators of Gaussian process components is discussed in detail. Even if constructed using Gaussian process building blocks, Bayesian field theories are typically non-Gaussian and have thus to be solved numerically. According to increasing computational resources the class of non-Gaussian Bayesian field theories of practical interest which are numerically feasible is steadily growing. Models which turn out to be computationally too demanding can serve as starting point to construct easier to solve parametric approaches, using for example variational techniques.