Compromise-free Bayesian neural networks

1 code implementation • 25 Apr 2020 • Kamran Javid, Will Handley, Mike Hobson, Anthony Lasenby

We conduct a thorough analysis of the relationship between the out-of-sample performance and the Bayesian evidence (marginal likelihood) of Bayesian neural networks (BNNs), as well as looking at the performance of ensembles of BNNs, both using the Boston housing dataset.

Nested sampling on non-trivial geometries

1 code implementation • 22 May 2019 • Kamran Javid

For parameters which represent points on a circle or torus, the trial distribution is `wrapped' around the domain of the posterior distribution such that samples cannot be rejected automatically when evaluating the Metropolis ratio due to being outside the sampling domain.

Computation Instrumentation and Methods for Astrophysics Data Analysis, Statistics and Probability