Department of Computer Science, UBC

Tue 19 Sep 2017, 12:30pm
Scientific Computation and Applied & Industrial Mathematics
ESB 4133 (PIMS Lounge)

Numerical Analysis in Visual Computing: not too little, not too much

ESB 4133 (PIMS Lounge)
Tue 19 Sep 2017, 12:30pm1:30pm
Abstract
Visual computing is a wide area that includes computer graphics and image processing, where the ``eyeballnorm'' rules.
I will briefly discuss two case studies involving numerical methods and analysis applied to this area. The first involves motion simulation and calibration of soft objects such as cloth, plants and skin. The governing elastodynamics PDE system, discretized in space already at the variational level using corotated FEM, leads to a large, expensive to assemble, dynamical system in time, where the damped motion may mask highly oscillatory stiffness. Geometric integration ideas are making their way into visual computing research these days in search for more quantitative computations.
The other case study involves some image processing problems where there is a premium for local approaches that do not necessarily use underlying PDEs. I will demonstrate and discuss.
The examples used are from several published or submitted papers.
hide

Technical University of Munich

Tue 19 Sep 2017, 3:00pm
SPECIAL
LSK 306

Overview of the Julia programming language: an 8 hour minicourse

LSK 306
Tue 19 Sep 2017, 3:00pm5:30pm
Details
Course description: The Julia programming language is designed to be a high level language for numerical computing, that is as fast as C or Fortran, despite employing a high level syntax. Since its first release in 2012 it has been continually improved and build a fast growing community around it.
The aim of this course is to give an overview of the key concepts of the Julia programming language as well as explain the advantages over other languages designed for numerical computing, as e.g. Matlab or R. Furthermore it demonstrates how readily available packages developed with Julia can be used to solve common problems occurring in numerical analysis, such as  linear systems of equations  non linear systems of equations  ordinary differential equations  linear programs. The course will also cover the basic tasks frequently encountered by numerical analysts: benchmarking, plotting and debugging. If time permits we will also explore possibilities for using Julia in deep learning applications.
hide

Tokyo Institute of Technology

Tue 19 Sep 2017, 3:30pm
Diff. Geom, Math. Phys., PDE Seminar
ESB 2012

On uniqueness for the harmonic map heat flow in supercritical dimensions

ESB 2012
Tue 19 Sep 2017, 3:30pm4:30pm
Abstract
We examine the question of uniqueness for the equivariant reduction of the harmonic map heat flow in the energy supercritical dimension. It is shown that, generically, singular data can give rise to two distinct solutions which are both stable, and satisfy the local energy inequality. We also discuss how uniqueness can be retrieved. This is a joint work with Pierre Germain and TejEddine Ghoul.
hide

University of Victoria

Wed 20 Sep 2017, 3:00pm
Probability Seminar
ESB 2012

A characterization of the GFF

ESB 2012
Wed 20 Sep 2017, 3:00pm4:00pm
Abstract
We characterize the GFF as the only random distribution which is conformally invariant and satisfies a domain Markov property. Joint work with Nathanael Berestycki and Ellen Powell.
hide

Technical University of Munich

Wed 20 Sep 2017, 3:00pm
SPECIAL
LSK 306

Overview of the Julia programming Language: an 8 hour minicourse Part II

LSK 306
Wed 20 Sep 2017, 3:00pm5:30pm
Details
Course description: The Julia programming language is designed to be a high level language for numerical computing, that is as fast as C or Fortran, despite employing a high level syntax. Since its first release in 2012 it has been continually improved and build a fast growing community around it.
The aim of this course is to give an overview of the key concepts of the Julia programming language as well as explain the advantages over other languages designed for numerical computing, as e.g. Matlab or R. Furthermore it demonstrates how readily available packages developed with Julia can be used to solve common problems occurring in numerical analysis, such as  linear systems of equations  non linear systems of equations  ordinary differential equations  linear programs. The course will also cover the basic tasks frequently encountered by numerical analysts: benchmarking, plotting and debugging. If time permits we will also explore possibilities for using Julia in deep learning applications.
hide

CINVESTAV

Wed 20 Sep 2017, 3:15pm
Topology and related seminars
ESB 4133 (PIMS Lounge)

Hopf invariants in the motion planning problem

ESB 4133 (PIMS Lounge)
Wed 20 Sep 2017, 3:15pm4:15pm
Abstract
Hopf invariants, a basic construction in homotopy theory, are closely related to Lusternik–Schnirelmann category which, in turn, can be defined as the sectional category of a certain evaluation map. In this talk I'll introduce the notion of Hopf invariants for general fibrations and exhibit a connection between the Hopf invariants for a product fibration and those for the factors. Applications will be drawn to the motion planning problem.
hide

Technical University of Munich

Thu 21 Sep 2017, 3:00pm
SPECIAL
LSK 306

Overview of the Julia programming language: an 8 hour minicourse Part III

LSK 306
Thu 21 Sep 2017, 3:00pm5:30pm
Details
Course description: The Julia programming language is designed to be a high level language for numerical computing, that is as fast as C or Fortran, despite employing a high level syntax. Since its first release in 2012 it has been continually improved and build a fast growing community around it.
The aim of this course is to give an overview of the key concepts of the Julia programming language as well as explain the advantages over other languages designed for numerical computing, as e.g. Matlab or R. Furthermore it demonstrates how readily available packages developed with Julia can be used to solve common problems occurring in numerical analysis, such as  linear systems of equations  non linear systems of equations  ordinary differential equations  linear programs. The course will also cover the basic tasks frequently encountered by numerical analysts: benchmarking, plotting and debugging. If time permits we will also explore possibilities for using Julia in deep learning applications.
hide

UBC Math

Fri 22 Sep 2017, 3:00pm
Department Colloquium
ESB 2012

Some directions in analysis and geometry of probability measures

ESB 2012
Fri 22 Sep 2017, 3:00pm4:00pm
Abstract
Probability measures are key objects in many scientific and engineering areas that deal with randomness, distributions, data sets, etc. When coupled with optimization, many interesting questions naturally arise. In this talk, I will explain a few of such questions from the point of view of optimal transport theory, which gives a natural and robust framework for studying probability measures. These involve among others, matching probability measures in an optimal way following certain rules (e.g martingale), as well as finding geometric averages between probability measures.
hide

Note for Attendees
Future Dates/Times of this ongoing minicourse: Wednesday, Sep 20, 3:005:30pm, Thursday, Sep 21 3:005:30pm.