Physical Sciences and Mathematics Commons^™

Investigating The Impact Of Unsupervised Feature-Extraction From Multi-Wavelength Image Data For Photometric Classification Of Stars, Galaxies And Qsos, Annika Lindh

Conference papers

Accurate classification of astronomical objects currently relies on spectroscopic data. Acquiring this data is time-consuming and expensive compared to photometric data. Hence, improving the accuracy of photometric classification could lead to far better coverage and faster classification pipelines. This paper investigates the benefit of using unsupervised feature-extraction from multi-wavelength image data for photometric classification of stars, galaxies and QSOs. An unsupervised Deep Belief Network is used, giving the model a higher level of interpretability thanks to its generative nature and layer-wise training. A Random Forest classifier is used to measure the contribution of the novel features compared to a set …

Factorized Runge-Kutta-Chebyshev Methods, Stephen O'Sullivan

Conference papers

The second-order extended stability Factorized Runge-Kutta-Chebyshev (FRKC2) class of explicit schemes for the integration of large systems of PDEs with diffusive terms is presented. FRKC2 schemes are straightforward to implement through ordered sequences of forward Euler steps with complex stepsizes, and easily parallelised for large scale problems on distributed architectures.

Preserving 7 digits for accuracy at 16 digit precision, the schemes are theoretically capable of maintaining internal stability at acceleration factors in excess of 6000 with respect to standard explicit Runge-Kutta methods. The stability domains have approximately the same extents as those of RKC schemes, and are a third longer …

Particle Transport And Stochastic Acceleration In The Giant Lobes Of Centaurus A, Stephen O'Sullivan

Conference papers

The conditions within the giant lobes of Centaurus A are reviewed in light of recent radio and γ‐ray observations. Data from WMAP and ground‐based telescopes in conjunction with measurements from Fermi‐LAT constrain the characteristic field strength and the maximum electron energy. The implications for the transport of energetic particles are discussed in terms of residence times and cooling times within the lobes. Acceleration of electrons and UHECR via the second order Fermi mechanism is discussed.

Relativistic Particle Acceleration In Tangled Magnetic Fields, Stephen O'Sullivan, Brian Reville, Andrew Taylor

Conference papers

We present simulations of the transport of fast particles through three-dimensional turbulent magnetic field configurations. A time dependency is imposed on the plane wave modes used in constructing these fields such than acceleration via the second-order Fermi process is possible. We consider simulations of models with low and high turbulence levels for non-relativistic waves. The predictions of quasi-linear theory are discussed with respect to the simulation data. We conclude that for pure stochastic acceleration via Alfvén waves to be plausible as the generator of UHECR in Cen A, the baryon number density would need to be several orders of magnitude …

Off To A Good Start: Using Clustering To Select The Initial Training Set In Active Learning, Rong Hu, Brian Mac Namee, Sarah Jane Delany

Conference papers

Active learning (AL) is used in textual classification to alleviate the cost of labelling documents for training. An important issue in AL is the selection of a representative sample of documents to label for the initial training set that seeds the process, and clustering techniques have been successfully used in this regard. However, the clustering techniques used are nondeterministic which causes inconsistent behaviour in the AL process. In this paper we first illustrate the problems associated with using non-deterministic clustering for initial training set selection in AL. We then examine the performance of three deterministic clustering techniques for this task …

Implications Of The Hall Effect For Turbulent Molecular Clouds, Turlough Downes, Stephen O'Sullivan

Conference papers

No abstract provided.

Numerical Modeling Of Weakly Ionized Plasmas, Stephen O'Sullivan, Turlough Downes

Conference papers

No abstract provided.

Reissner–Nordstrom¨ Expansion, Emil Prodanov, Rossen Ivanov, V.G. Gueorguiev

Conference papers

We propose a classical mechanism for the cosmic expansion during the radiation-dominated era, assuming the Universe as a two-component gas. The first component is the ultra-relativistic “standard” fraction described by an equation of state of an ideal quantum gas of massless particles. The second component consist of superheavy charged particles and their interaction with the “standard” fraction drives the expansion. This interaction is described by the Reissner–Nordstr¨om metric purely geometrically — the superheavy charged particles are modeled as zero-dimensional naked singularities which exhibit gravitational repulsion. The radius of a repulsive sphere, surrounding a naked singularity of charge Q, is inversely …

The Nonlinear Amplification Of Magnetic Fields By Cosmic Rays At Supernova Remnant Shocks, Brian Reville, Peter Duffy, Stephen O'Sullivan

Conference papers

A central problem in the theory of cosmic ray acceleration at supernova shock fronts is the generation of turbulent magnetic fields needed to scatter particles across a shock front. In this paper we build on previous studies [1], [2] into the effect of streaming cosmic rays produced by the outer shocks of supernova remnants on the stochastic component of the magnetic field. A three dimensional, MHD code has been constructed which demonstrates the nonlinear growth of the turbulent field.