1 code implementation • 6 Dec 2023 • Pablo Lemos, Nikolay Malkin, Will Handley, Yoshua Bengio, Yashar Hezaveh, Laurence Perreault-Levasseur
We present a performant, general-purpose gradient-guided nested sampling algorithm, ${\tt GGNS}$, combining the state of the art in differentiable programming, Hamiltonian slice sampling, clustering, mode separation, dynamic nested sampling, and parallelization.
1 code implementation • 7 Nov 2023 • Namu Kroupa, David Yallup, Will Handley, Michael Hobson
Using a fully Bayesian approach, Gaussian Process regression is extended to include marginalisation over the kernel choice and kernel hyperparameters.
1 code implementation • 4 May 2023 • Harry Bevins, Will Handley, Thomas Gessey-Jones
We demonstrate the performance of the piecewise flows using some standard benchmarks and compare the accuracy of the flows to the approach taken in Stimper et al. (2022) for modelling multi-modal distributions.
no code implementations • 23 May 2022 • David Yallup, Will Handley, Mike Hobson, Anthony Lasenby, Pablo Lemos
The true posterior distribution of a Bayesian neural network is massively multimodal.
1 code implementation • 24 Feb 2021 • Justin Alsing, Will Handley
In this letter we show that parametric bijectors trained on samples from a desired prior density provide a general-purpose method for constructing transformations from the uniform base density to a target prior, enabling the practical use of nested sampling under arbitrary priors.
1 code implementation • 12 Jan 2021 • The DarkMachines High Dimensional Sampling Group, Csaba Balázs, Melissa van Beekveld, Sascha Caron, Barry M. Dillon, Ben Farmer, Andrew Fowlie, Will Handley, Luc Hendriks, Guðlaugur Jóhannesson, Adam Leinweber, Judita Mamužić, Gregory D. Martinez, Pat Scott, Eduardo C. Garrido-Merchán, Roberto Ruiz de Austri, Zachary Searle, Bob Stienen, Joaquin Vanschoren, Martin White
Optimisation problems are ubiquitous in particle and astrophysics, and involve locating the optimum of a complicated function of many parameters that may be computationally expensive to evaluate.
Bayesian Optimisation High Energy Physics - Phenomenology Computational Physics
1 code implementation • 7 Sep 2020 • The GAMBIT Cosmology Workgroup, :, Janina J. Renk, Patrick Stöcker, Sanjay Bloor, Selim Hotinli, Csaba Balázs, Torsten Bringmann, Tomás E. Gonzalo, Will Handley, Sebastian Hoof, Cullan Howlett, Felix Kahlhoefer, Pat Scott, Aaron C. Vincent, Martin White
We introduce $\sf{CosmoBit}$, a module within the open-source $\sf{GAMBIT}$ software framework for exploring connections between cosmology and particle physics with joint global fits.
Cosmology and Nongalactic Astrophysics High Energy Physics - Phenomenology
1 code implementation • 7 Sep 2020 • The GAMBIT Cosmology Workgroup, :, Patrick Stöcker, Csaba Balázs, Sanjay Bloor, Torsten Bringmann, Tomás E. Gonzalo, Will Handley, Selim Hotinli, Cullan Howlett, Felix Kahlhoefer, Janina J. Renk, Pat Scott, Aaron C. Vincent, Martin White
We determine the upper limit on the mass of the lightest neutrino from the most robust recent cosmological and terrestrial data.
Cosmology and Nongalactic Astrophysics High Energy Physics - Phenomenology
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.
1 code implementation • 12 Sep 2018 • Edward Higson, Will Handley, Michael Hobson, Anthony Lasenby
Our approach can also be readily applied to neural networks, where it allows the network architecture to be determined by the data in a principled Bayesian manner by treating the number of nodes and hidden layers as parameters.
3 code implementations • 16 Apr 2018 • Edward Higson, Will Handley, Mike Hobson, Anthony Lasenby
Nested sampling is an increasingly popular technique for Bayesian computation - in particular for multimodal, degenerate and high-dimensional problems.
Computation Cosmology and Nongalactic Astrophysics Instrumentation and Methods for Astrophysics Data Analysis, Statistics and Probability
2 code implementations • 11 Apr 2017 • Edward Higson, Will Handley, Mike Hobson, Anthony Lasenby
We introduce dynamic nested sampling: a generalisation of the nested sampling algorithm in which the number of "live points" varies to allocate samples more efficiently.
Computation Instrumentation and Methods for Astrophysics Data Analysis, Statistics and Probability Methodology
2 code implementations • 28 Mar 2017 • Edward Higson, Will Handley, Mike Hobson, Anthony Lasenby
Sampling errors in nested sampling parameter estimation differ from those in Bayesian evidence calculation, but have been little studied in the literature.
Methodology Instrumentation and Methods for Astrophysics Applications