Dr Peter Loxley
Lecturer - School of Science and Technology
Phone: +61 2 6773 2307
My PhD was on the field theory of solitons, and the statistical mechanics of nucleation in the ferromagnetic spin-chain model. This work was later used in the design of new types of high-density magnetic storage media for computer hard disks.
My first postdoctoral position was in mathematical neuroscience, in the School of Physics at the University of Sydney. I investigated nonlinear neural dynamics, and proposed a model where neural competition leads to a nonlinear wave instability. This model instability quantifies how a visual stimulus can lead to a visual percept known as binocular rivalry.
My second postdoctoral position was in the Center for Nonlinear Studies at Los Alamos National Laboratory. Here, amongst other things, I worked on a statistical model of sensory processing, similar to independent component analysis, that describes how simple cells in the primary visual cortex may reduce redundancy in natural sensory data. I also investigated bi-stable steady states in two-dimensional turbulence by considering maximum entropy states in the point-vortex model. This work was motivated by attempting to understand the dynamics of bi-stable ocean currents such as the Kuroshio current along the east coast of Japan.
I was then an adjunct lecturer at the University of New Mexico (in Los Alamos) teaching first-year maths and computer science, before joining UNE.
PhD (Physics) University of Western Australia
BSc Hons 1st class (Physics) University of Sydney
Discrete Mathematics, Data Structures and Algorithms, Algorithms in Machine Learning, Reinforcement Learning.
Primary Research Area/sStatistical physics; Probabilistic models in machine learning; Information theory and neural computation; Nonlinear dynamics and pattern formation