Research

I am a research scientist at CSIRO Mathematics, Informatics and Statistics in Perth, Australia. Previously I was a PhD student in machine learning at the School of Informatics, University of Edinburgh, after completing my bachelor’s degree in software engineering at the Australian National University.

My current work is in statistical and computational methods for inference in nonlinear and continuous-time state space models. This includes aspects of sequential Monte Carlo, Markov chain Monte Carlo, Bayesian filtering, stochastic and ordinary differential equations. On the computational side it includes general purpose GPU programming (GPGPU) and high performance computing. Environmental applications are the main motivation, particularly marine biogeochemistry. I have previously worked on applications in Functional Magnetic Resonance Imaging (fMRI).

Contact lawrence.murray@csiro.au.

Publications

2012

• Jones, E.M.; Oke, P.R.; Rizwi, F. & Murray, L.M. (2012). Assimilation of glider and mooring data into a coastal ocean model. Ocean Modelling, to appear.
• Murray, L.M. (2012). GPU acceleration of Runge-Kutta integrators. IEEE Transactions on Parallel and Distributed Systems. 23, 94-101. [full text]

2011

• Domanski, L.; Bednarz, T.; Gureyev, T.; Murray, L.M.; Huang, E. & Taylor, J. (2011). Applications of Heterogeneous Computing in Computational and Simulation Science. 1st International Workshop on Cloud Computing and Scientific Applications.
• Nandapalan, N.; Brent, R.; Murray, L.M. and Rendell, A. (2011). High-performance Pseudo-Random Number Generation on Graphics Processing Units. 9th International Conference on Parallel Processing and Applied Mathematics. [arXiv]
• Murray, L.M. and Storkey, A. (2011). Particle smoothing in continuous time: A fast approach via density estimation. IEEE Transactions on Signal Processing, 59, 1017-1026. [full text]
• Murray, L.M. (2011). GPU acceleration of the particle filter: The Metropolis resampler. Distributed machine learning and sparse representation with massive data-sets (DMMD 2011).

2010

• Murray, L.M. (2010). Distributed Markov chain Monte Carlo. NIPS 2010 workshop: Learning on Cores, Clusters and Clouds.
• Murray, L.M.; Jones, E; Parslow, J.; Campbell, E and Margvelashvili, N. (2010). Discussion item on Andrieu, C.; Doucet, A. and Holenstein, R. (2010) Particle Markov chain Monte Carlo methods. Journal of the Royal Statistical Society Series B, 72, 269-302.
• Jones, E.; Parslow, J. and Murray, L.M. (2010) A Bayesian approach to state and parameter estimation in a Phytoplankton-Zooplankton model. Australian Meteorological and Oceanographic Journal, 59, 7-16. [full text]

2007-2009

• Murray, L.M. (2009) Bayesian Learning of Continuous Time Dynamical Systems (with applications in Functional Magnetic Resonance Imaging). PhD thesis. [full text]
• Murray, L.M. and Storkey, A.J. (2008) Continuous Time Particle Filtering for fMRI. Advances in Neural Information Processing Systems, 20, 1049-1068. [full text] [poster]
• Storkey, A.J., Simonotto, E., Whalley, H., Lawrie, S., Murray, L.M. and McGonigle, D. (2007) Learning structural equation models for fMRI. Advances in Neural Information Processing Systems, 19, 1329-1336.

Software

The dysii Dynamic Systems Library is a direct product of my research work to date.

I’ve posted the final version of my PhD thesis, "Bayesian Learning of Continuous-Time Dynamical Systems, with Applications in Functional Magnetic Resonance Imaging" to the research page. Now assessed, corrected and passed! Note that this may serve as a useful manual for some of the detail behind the algorithms of the dysii Dynamic Systems Library.

Version 1.4.0 of the dysii Dynamic Systems Library has been released. This is a major new release with a number of additional features and performance enhancements, as well as representing a consolidation of code and maturation of much of the API. Particular new features include: The kernel forward-backward and two-filter smoothers, suitable for fast, large-scale [...]

I’ve made my PhD thesis available on the research page, "Bayesian Learning of Continuous Time Dynamical Systems (with Applications in Functional Magnetic Resonance Imaging)". The thesis considers Bayesian filtering and smoothing for state and parameter estimation in general non-linear, non-Gaussian systems using stochastic differential models. It is the theoretical basis for impending updates to the [...]