research

I’ve added one preprint and one older workshop paper to arXiv.org, given recent interest, see below.

Murray, L. M.; Jones, E. M. & Parslow, J. (2012). On collapsed state-space models and the particle marginal Metropolis-Hastings sampler. In review. [arXiv]

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). [arXiv]

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 approximate inference in continuous-time stochastic models, as documented in my recent PhD thesis.
• Overhauled kd tree implementation, featuring distributed partitioning, dual-tree and self-tree evaluations, particularly useful for the new smoothers above.
• Improved stochastic Runge-Kutta and new Euler-Maruyama method for integrating stochastic differential equations.
• Performance improvements resulting from continued profiling, including more aggressive inlining and less dependence on virtuals.
• A new installation guide, available in the INSTALL.txt file of the distribution. Also note that with Boost 1.35 now released, dysii no longer requires the latest CVS of Boost, making it much simpler to install.

Full details are included in the VERSION.txt file of the distribution.

A couple of examples of applications using dysii are expected to be released within a matter of days also. These should provide an excellent starting point for those wishing to use the library for their own work.

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 dysii Dynamic Systems Library, including the kernel forward-backward and kernel two-filter smoothers, and distributed implementation of particle filters and kd trees.

I should note that this should be considered a draft version, as it is yet to be examined, and corrections may need to be made after it is. I’m providing it here mainly for the purposes of documenting dysii at this stage.