A Non-Parametric Bayesian Method for Inferring Hidden Causes

TL;DR

This paper introduces a non-parametric Bayesian method for inferring hidden causes in data, allowing for an unbounded number of potential causes and using Gibbs sampling for structure learning, demonstrated on simulated and real medical data.

Contribution

It proposes a novel non-parametric Bayesian approach that simplifies structure learning of hidden causes by assuming a finite influence set, enabling Gibbs sampling instead of complex algorithms.

Findings

Effective in discovering hidden causes in simulated data

Successfully applied to real medical dataset

Outperforms traditional methods in flexibility and inference quality

Abstract

We present a non-parametric Bayesian approach to structure learning with hidden causes. Previous Bayesian treatments of this problem define a prior over the number of hidden causes and use algorithms such as reversible jump Markov chain Monte Carlo to move between solutions. In contrast, we assume that the number of hidden causes is unbounded, but only a finite number influence observable variables. This makes it possible to use a Gibbs sampler to approximate the distribution over causal structures. We evaluate the performance of both approaches in discovering hidden causes in simulated data, and use our non-parametric approach to discover hidden causes in a real medical dataset.