What's On at UWA

Pablo Spiga (Universita' di Milano-Bicocca)

will speak on

Have we ever tried to count Cayley graphs?

at 3pm Friday March the 21st in Weatherburn Lecture Theatre.

Abstract:

In this talk we give some elementary upper bounds on the number of finite Cayley graphs. The asymptotic number of Cayley graphs is much harder to pin down and we give a brief outline of the main technique and the main ingredients needed for this counting. On the way we leave some problems on oriented Cayley graphs and tournaments.

In the week beginning 24 March 2014, iVEC will offer the following short courses on supercomputing topics:

Introduction to iVEC: 1:00 – 2:00pm Mon 24th March

Introduction to Linux: 3:00 – 4:00pm Mon 24th March

Introduction to Supercomputing: 10:00am – 4:00pm Tues 25th March

Developing with MPI and OpenMP: 10:00am – 4:00pm Wed 26th March

Debugging with DDT: 10am – 12:00 noon Thurs 27th March

Introduction to Optimisation and Profiling: 1:00 – 4:00pm Thurs 27 March

Courses are delivered in a face to face classroom style. Attendees are encouraged to bring and work on their own laptops. Staff from the Supercomputing Team will be facilitating so you can meet and chat with them.

Free parking available on-site.

Courses are free of charge and open to all, however places are limited.

Light refreshments and lunch will be provided on each day (no lunch on Monday).

Register online at http://www.ivec.org/ai1ec_event/ivec-supercomputing-training-february-2014-2/?instance_id=

Michael Zieve (University of Michigan)

will speak on

Primitive groups, diophantine equations, and functional equations

at 3pm Friday March the 28th in Weatherburn Lecture Theatre.

Abstract:

I will explain how results about primitive groups play a crucial role in proving results about diophantine equations and functional equations. A sample application is that, for any polynomial f(X) with rational coefficients, the function Q-->Q defined by c --> f(c) is (<=6)-to-1 over all but finitely many values.

Cai-Heng Li (UWA)

will speak on

Edge transitive dessins d'enfant

at 3pm Friday April the 4th in Weatherburn Lecture Theatre.

Abstract:

A 2-cell embedding of a bipartite graph in an orientable surface is called a dessin d'enfant. Thus a dessin d'enfant is an orientable bipartite map. I will present an explicit representation of an edge transitive dessin in terms of a group with two distinguished generators, and apply it to study the dessin.

There will be presentations by international visitors, games, prizes, a photo booth, treats on offer and lots and lots of information about exchange and study abroad opportunities for UWA students.

Come along to Profectus UWA's Inaugral 2014 AGM on Tuesday April 8. If you are passionate about business and entrepreneurship be sure to drop by at 1pm to learn more about our agenda. We will answer any questions you have about Profectus and events being held this semester!!!

Steven Noble (Brunel University)

will speak on

Graphs are to matroids what ribbon graphs are to ...?

at 3pm Friday April the 11th in Weatherburn Lecture Theatre.

Abstract:

Much of the combinatorial structure of an abstract connected graph is encoded in its set of spanning trees. These form one of the canonical examples of the bases of a matroid. Ribbon graphs contain extra topological information on the embedding of the underlying abstract graph in a surface. The structures playing the role of spanning trees are the subgraphs having one boundary component. The combinatorial structure which they form is a delta-matroid, which roughly speaking is like a matroid except that the bases do not need to have the same size. We will discuss three classes of delta-matroids, some natural operations on ribbon graphs which correspond to natural operations on delta-matroids, give an excluded minor theorem for one of the classes of delta-matroids and, time-permitting describe the Bollobas-Riordan polynomial which forms one way of generalizing the Tutte polynomial to ribbon graphs.

Michael Giudici (UWA)

will speak on

Some recent results on elusive groups

at 3pm Friday May the 2nd in Weatherburn Lecture Theatre.

Abstract:

Every transitive permutation group has a derangement of prime power order but not necessarily a derangement of prime order. A transitive permutation group is called elusive if it has no derangements of prime order. I will talk about some recent results on this topic including elusive groups of automorphisms of graphs of small valency and a new construction of elusive groups.

Luke Morgan (UWA)

will speak on

Locally-transitive graphs and their vertex stabilisers

at 3pm Friday May the 9th in Weatherburn Lecture Theatre.

Abstract:

Let D be a graph with a group of automorphisms G and suppose that D is G-locally-arc-transitive (for each vertex x of D the vertex stabiliser G(x) acts transitively on the neighbourhood of x). Fixing the valency of D one can ask if there is a bound on the order of G(x)? For valency three there are fundamental results due to Tutte and Goldschmidt. We'll instead fix the `local actions’, that is, the possible permutation groups induced on neighbourhoods in D (there can be at most two of these) and ask again: is there a bound on the order of G(x)? I will revisit some of the interesting results on this question. Then I will talk about some recent results like: when there can be a bound; when there is no bound; given a bound (and a little more) an instance when G(x) can be completely determined. Spread throughout this is joint work with Giudici, Giudici-Ivanov-Praeger and Spiga-Verret.

Denis Weaire

Gledden Visiting Fellow, Institute for Advanced Studies, UWA

and

School of Physics, Trinity College, Dublin, Ireland

will speak on

Classic problems of packing in 2d, 3d and on a cylinder

at 3pm Friday May the 16th in Weatherburn Lecture Theatre.

Abstract:

Some classic problems of optimal packing are reviewed: the Kepler Problem, the Kelvin Problem, and cylindrical packings of spheres and disks. The latter are tentatively associated with the celebrated phenomenon of spiral Phyllotaxis, much in evidence in the splendid gardens of UWA. If the speaker omits or forgets anything, it is to be found in the book Pursuit of Perfect Packing, T. Aste and D. Weaire, 2nd Edition.

iVEC and CSIRO invite you to attend training on MOOSE – Multiphysics Object-Oriented Simulation Environment. The MOOSE platform is a general problem solver that can accommodate many mathematical models. It has recently been released as open-source software, making it even easier for academic researchers to adopt. Some of the capabilities of MOOSE include:

Fully-coupled, fully-implicit multiphysics solver

Dimension independent physics

Automatically parallel (largest runs >100,000 CPU cores!)

Modular development simplifies code reuse

Built-in mesh adaptivity

Continuous and Discontinuous Galerkin (DG) (at the same time!)

Intuitive parallel multi scale solves

Dimension agnostic, parallel geometric search (for contact related applications)

Flexible, plugable graphical user interface

~30 plugable interfaces allow specialisation of every part of the solve

More information about MOOSE can be found at the MOOSE website: http://mooseframework.org

Please refer to the URL for the Program Overview and more information.

Cody Permann’s visit is sponsored by CSIRO to support researchers in WA. If you would like to arrange a meeting with Cody, please contact Thomas Poulet <[email protected]>, other enquiries should be directed to Rebecca Hartman-Baker <[email protected]>.

Although the training is free, you must register your attendance (through the URL) as places are limited. The intensive program builds a solid foundation for MOOSE use, so attendance on all three days is advised.

Gordon Royle (UWA)

will speak on

Chromatic roots of graphs and matroids

at 3pm Friday May the 23rd in Weatherburn Lecture Theatre.

Abstract:

The location of the real and/or complex roots of the chromatic polynomial of a graph has been studied for many years, both by combinatorial mathematicians and statistical physicists, yet despite this many fundamental questions remain unsolved. And even though the chromatic polynomial is most generally a matroidal concept, very little indeed is known about the location of chromatic roots of matroids that are not graphic or cographic. In this talk, I will present a necessarily-personal survey of the major results and my favourite open problems in this area.

Simon Smith (UWA)

will speak on

Taking products of permutation groups

at 3pm Friday May the 30th in Weatherburn Lecture Theatre.

Abstract:

At the AustMS conference, I introduced a new product for permutation groups. At the time, I was in a rush to announce the result because it could be used to solve an open question of P. E. Caprace and N. Monod. Since that talk, much has changed: I have an entirely new construction method that expands the scope of the construction, and I have extended many of the results describing its properties.

In this talk I'll introduce the product again, but this time in more detail. I'll sketch a proof of one of its most striking properties, and I'll talk about three areas of research in which the product has proved to be important (one of which is my DECRA project!).

The content of the talk will be largely combinatorial, and there will be no topology (although I may use the word "closed" accidentally, for which I hope I will be forgiven).

The 2014 Blakers Mathematics Competition is now open.

Eligible candidates are all undergraduates in first, second and third years at a WA university.

Entries must be received by Friday, 26 September, 2014. They may be mailed or given to Dr Greg Gamble, School of Mathematics and Statistics, The University of Western Australia, Crawley, 6009. It is recommended that a PDF-scanned copy of your entry is also emailed to [email protected]. Please mark your entry Blakers Mathematics Competition 2014, and include your name, address, email address, university, and number of years you have been attending any tertiary institution.

More details about how to enter can be found at http://www.maths.uwa.edu.au/students/competitions

In the week beginning 2 June 2014, iVEC will offer the following free short courses on supercomputing topics:

Introduction to iVEC: 2:00 – 3:00pm Mon 2nd June

Introduction to Linux: 3:30 – 4:30pm Mon 2nd June

Introduction to Supercomputing: 10:00am – 4:00pm Tues 3rd June

Developing with MPI and OpenMP: 10:00am – 4:00pm Wed 4th June.

Further details of the courses and the registration form can be found on the iVEC website. Courses are delivered in a face to face classroom style. Attendees are encouraged to bring and work on their own laptops. Staff from the Supercomputing Team will be facilitating so you can meet and chat with them.

Any queries, please contact Dr Rebecca Hartman-Baker – [email protected]

This is a free event however, you must register as places are limited.

Frank Lübeck (RWTH Aachen)

will speak on

Computing Kazhdan-Lusztig Polynomials and some Applications

at 3pm Thursday June the 5th in Maths Lecture Room 2.

Abstract:

My interest in Kazhdan-Lusztig polynomials comes from a certain character formula for reductive algebraic groups, which was first conjectured by Lusztig.

More generally, parabolic Kazhdan-Lusztig polynomials can be defined for arbitrary Coxeter groups. These polynomials are notoriously difficult to compute. I will report on an implementation of an algorithm to compute them, which goes much further than previous programs.

It turned out that some of the coefficients of Kazhdan-Lusztig polynomials that I was able to compute have interesting (and for me unexpected) interpretations in the context of conjectures by Guralnick (on a bound of the dimension of first cohomology groups for finite groups) and Wall (on the number of maximal subgroups of any finite group). I will also explain these conjectures and sketch the connection between these topics.

The aim of this course is to introduce you to basic statistics. It will cover descriptive statistics (means and standard deviations); data exploration; basic categorical data analysis; simple linear regression and basic analysis of variance (ANOVA). The statistical package SPSS will be used to illustrate the ideas demonstrated. The course will be held in a computer laboratory allowing participants to immediately apply the material covered through a series of practical examples.

R is a free and extremely powerful language and software environment for statistical computing, data analysis, and graphics. The course is designed for those who have no experience with R, but have a basic understanding of statistics. The course will include: Introduction to R: How to install R on your computer; basic R commands, how to use and understand the R help pages. Data: Reading in data and data manipulation; summarising data; basic statistical analysis and fitting linear models. Graphics and output: Basic plotting commands and how to customise your plots; how to export your plots and output in a user-friendly format. Functions: Writing simple functions and flow control structures.

The course is designed for people with knowledge of basic statistics who want to learn more about regression and analysis of variance (ANOVA).

This course covers techniques that can be used to analyse data with continuous and categorical variables. The course will begin with simple linear regression and then proceed with approaches that can be used with more than two variables such as multiple regression. ANOVA with interactions and blocking will also be covered. The course will end with techniques that address the analysis of binary or ordinal variables.

The course is designed as an applied course in Structural Equation Modelling (SEM) using the Mplus software package. SEM is used widely by researchers to test complex relationships among observed (measured) and latent (unobserved) variables and subsumes other analytical techniques such as regression, path analysis, factor analysis, and canonical correlation. Mplus is rapidly becoming the program of choice for the analysis of SEMs. Mplus offers a general modelling framework that allows both the modelling of cross-sectional and longitudinal data using observed variables that are a combination of continuous and categorical variables. In addition, Mplus analyses multilevel modelling structures.

The first three days of the course will be an introduction to SEM and the Mplus program. The focus of the last two days of the course is on the analysis of more advanced SEM models.

If you are familiar with the Mplus program and have an understanding of material typically covered in an introduction to SEM course, you may choose to attend only the last two days of the course.

If you have completed an introductory course in SEM using another program (e.g., Amos, Lisrel, EQS) but have not previously used the Mplus program, you may choose to attend the first day and then the last two days of the course.