EPFL (Ecole Polytechnique Federale de Lausanne)

The development of reduced order models for complex applications, offering the promise for rapid and accurate evaluation of the output of complex models under parameterized variation, remains a very active research area. Applications are found in problems which require many evaluations, sampled over a potentially large parameter space, such as in optimization, control, uncertainty quantification and applications where near real-time response is needed.

However, many challenges remain to secure the flexibility, robustness, and efficiency needed for general large-scale applications, in particular for nonlinear and/or time-dependent problems.

After giving a brief general introduction to reduced order models, we discuss developments in two different directions. In the first part, we discuss recent developments of reduced methods that conserve chosen invariants for nonlinear time-dependent problems. We pay particular attention to the development of reduced models for Hamiltonian problems and propose a greedy approach to build the basis. As we shall demonstrate, attention to the construction of the basis must be paid not only to ensure accuracy but also to ensure stability of the reduced model. Time permitting, we shall also briefly discuss how to extend the approach to include more general dissipative problems through the notion of port-Hamiltonians, resulting in reduced models that remain stable even in the limit of vanishing viscosity and also touch on extensions to Euler and Navier-Stokes equations.

The second part of the talk discusses the combination of reduced order modeling for nonlinear problems with the use of neural networks to overcome known problems of on-line efficiency for general nonlinear problems. We discuss the general idea in which training of the neural network becomes part of the offline part and demonstrate its potential through a number of examples, including for the incompressible Navier-Stokes equations with geometric variations.

This work has been done with in collaboration with B.F. Afkram (EPFL, CH), N. Ripamonti EPFL, CH) and S. Ubbiali (USI, CH).

The Ising model is one of the most classical statistical mechanics model, which has seen spectacular mathematical and physical developments for almost a century. The description of its scaling limit at the phase transition is at the center of a fascinating (conjectured) connection between statistical mechanics and field theories. I will discuss how recent mathematical progress allows one to make the connection between the two-dimensional Ising model and Conformal Field Theory rigorous. If time allows, I will discuss the insight this gives one into related models and field theories.

Based off joint works with S. Benoist, D. Chelkak, H. Duminil-Copin, R. Gheissari, K. Izyurov, F. Johansson-Viklund, K. Kytölä, S. Park and S. Smirnov

Gauss was the first to give a formula for the number of monic irreducible polynomials of degree n over a finite field. A natural problem is to determine the number of such polynomials for which certain coefficients are prescribed. While some asymptotic and existence results have been obtained, very few exact results are known. In this talk I shall present an algorithm which for any finite field GF(q) of characteristic p expresses the number of monic irreducibles of degree n for which the first l < p coefficients are prescribed, for n >= l and coprime to p, in terms of the number of GF(q^n)-rational points of certain affine varieties defined over GF(q). 
The GF(2) base field case is related to the distribution of binary Kloosterman sums, which have numerous applications in coding theory and cryptography, for example via the construction of bent functions. Using a variant of the algorithm, we present varieties (which are all curves) for l <= 7 and compute explicit formulae for l <= 5; before this work such formulae were only known for l <= 3. While this connection motivates the problem, the talk shall focus mainly on computational algebraic geometry, with the algorithm, theoretical questions and computational challenges taking centre stage.

Given a family $F$ of elliptic curves defined over $Q$, we are interested in the set $H(Y)$ of curves $E$ in $F$, of positive rank, and for which the minimum of the canonical heights of non-torsion rational points on $E$ is bounded by some parameter $Y$. When one can show that this set is finite, it is natural to investigate statistical properties of arithmetic objects attached to elliptic curves in the set $H(Y)$. We will describe some problems related to this, and will state some results in the case of families of quadratic twists of a fixed elliptic curve.

The vanishing mean curvature flow in Minkowski space is the

natural evolutionary generalisation of the minimal surface equation,

and has applications in cosmology as a model equation for cosmic

strings and membranes. The equation clearly admits initial data which

leads to singularity formation in finite time; Nguyen and Tian have

even shown stability of the singularity formation in low dimension. On

the other hand, Brendle and Lindblad separately have shown that all

"nearly flat" initial data leads to global existence of solutions. In

this talk, I describe an intermediate regime where global existence

of solutions can be proven on a codimension 1 set of initial data; and

where the codimension 1 condition is optimal --- The

catenoid, being a minimal surface in R^3, is a static solution to the

vanishing mean curvature flow. Its variational instability as a

minimal surface leads to a linear instability under the flow. By

appropriately "modding out" this unstable mode we can show the

existence of a stable manifold of initial data that gives rise to

solutions which scatters toward to the

catenoid. This is joint work with Roland Donninger, Joachim Krieger,

and Jeremy Szeftel. The preprint is available at http://arxiv.org/abs/1310.5606v1