The MoCaO Lecture Series will run daily from 25–29 August, from 11:00 AM to 12:00 PM AEST.
Speakers
Professor Tim Moroney is Professor in Applied and Computational Mathematics within QUT’s School of Mathematical Sciences. Among all the topics that have played a role throughout his research career, the one that features most prominently is numerical linear algebra, and particularly methods for sparse matrices. His current research interests come from various applications across water waves, droplets, surface reconstruction, moving interface problems, and fractional calculus.
Dr Qianqian Yang is a Senior Lecturer in Applied and Computational Mathematics within QUT’s School of Mathematical Sciences. Qianqian has extensive experience in developing computationally efficient methods for solving fractional order partial differential equations. With this background, her recent research interest lies in the application of these fractional calculus models to real-world problems, especially in the area of combining numerical simulations, fractional order models and magnetic resonance imaging (MRI) to probe in vivo brain tissue microstructure.
Abstract
This series of lectures will introduce the theoretical and practical ideas for iterative methods applied to sparse matrices. Sparse matrices arise in many applications across science, engineering, statistics, business and beyond. Iterative methods that exploit the sparsity of these matrices are essential for overcoming the scaling on storage and floating-point calculations that otherwise renders problems even of modest dimensionality impractical to solve.
A remarkably versatile family of numerical methods called Krylov subspace methods can be applied to a wide class of sparse matrix problems. In doing so, they impose only minimal requirements on the means by which a matrix is utilised, paving the way for many of today’s high-performance codes. This course covers Krylov subspace methods for two common problems: eigenvalue problems and linear systems, from their derivation through to efficient numerical implementation.
The lectures will be running in zoom.
Join Zoom Meeting
https://qut.zoom.us/j/87221801096?pwd=TGQK2WysRMquyaohaNhK9FryY3X5MC.1
Meeting ID: 872 2180 1096
Passcode: 519836
The lecture recordings will be available via our YouTube channel.
We are looking forward to seeing you soon.
Nadia (nsukhorukova@swin.edu.au)
Monday lecture: slides and code (PDF format only).
Tuesday lecture: slides and code files (PDF format only): Arnoldi Convergence, Arnoldi Theory Check, Arnoldi VS Naive and Gram-Schmidt Comparison
Wednesday lecture: slides
Thursday lecture: slides
Friday lecture: slides