Radford Neal's Research Interests

An approach to statistics in which all forms of uncertainty are expressed in terms of probability.

A way of computing high-dimensional integrals that is crucial for doing Bayesian inference.

Statistical models inspired by the way learning and computation may occur in the brain.

Statistical models phrased in terms of entities that we have invented to explain patterns we see in observable variables.

Efficient and accurate computation for statistical and scientific applications.

Ways of telling which methods for learning from data really work.

I have worked on various statistical applications, mostly of a biological nature.

Using models for data to find a compressed representation of it.

Representing information in a redundant form that allows errors to be corrected with high probability.

The quest for a fundamental understanding of how we can come to justified conclusions about the world.