The signature of a parametric curve is a sequence of tensors whose entries are iterated integrals, and they are central to the theory of rough paths in stochastic analysis.  For some special families of curves, such as polynomial paths and piecewise-linear paths, their parametrized signature tensors trace out algebraic varieties in the space of all tensors. We introduce these varieties and examine their fundamental properties, while highlighting their intimate connection to the problem of recovering a path from its signature. This is joint work with Peter Friz and Bernd Sturmfels. 

 We provide in this work a robust solution theory for random rough differential equations of mean field type

$$

dX_t = V\big( X_t,{\mathcal L}(X_t)\big)dt + \textrm{F}\bigl( X_t,{\mathcal L}(X_t)\bigr) dW_t,

$$

where $W$ is a random rough path and ${\mathcal L}(X_t)$ stands for the law of $X_t$, with mean field interaction in both the drift and diffusivity. Propagation of chaos results for large systems of interacting rough differential equations are obtained as a consequence, with explicit convergence rate. The development of these results requires the introduction of a new rough path-like setting and an associated notion of controlled path. We use crucially Lions' approach to differential calculus on Wasserstein space along the way. This is a joint work with I. Bailleul and R. Catellier.

Joint work with I. Bailleul (Rennes) and R. Catellier (Nice)

Abstract:
We present a numerical investigation of stochastic transport for the damped and driven incompressible 2D Euler fluid flows. According to Holm (Proc Roy Soc, 2015) and Cotter et al. (2017), the principles of transformation theory and multi-time homogenisation, respectively, imply a physically meaningful, data-driven approach for decomposing the fluid transport velocity into its drift and stochastic parts, for a certain class of fluid flows. We develop a new methodology to implement this velocity decomposition and then numerically integrate the resulting stochastic partial differential equation using a finite element discretisation. We show our numerical method is consistent.
Numerically, we perform the following analyses on this velocity decomposition. We first perform uncertainty quantification tests on the Lagrangian trajectories by comparing an ensemble of realisations of Lagrangian trajectories driven by the stochastic differential equation, and the Lagrangian trajectory driven by the ordinary differential equation. We then perform uncertainty quantification tests on the resulting stochastic partial differential equation by comparing the coarse-grid realisations of solutions of the stochastic partial differential equation with the ``true solutions'' of the deterministic fluid partial differential equation, computed on a refined grid. In these experiments, we also investigate the effect of varying the ensemble size and the number of prescribed stochastic terms. Further experiments are done to show the uncertainty quantification results "converge" to the truth, as the spatial resolution of the coarse grid is refined, implying our methodology is consistent. The uncertainty quantification tests are supplemented by analysing the L2 distance between the SPDE solution ensemble and the PDE solution. Statistical tests are also done on the distribution of the solutions of the stochastic partial differential equation. The numerical results confirm the suitability of the new methodology for decomposing the fluid transport velocity into its drift and stochastic parts, in the case of damped and driven incompressible 2D Euler fluid flows. This is the first step of a larger data assimilation project which we are embarking on. This is joint work with Colin Cotter, Dan Crisan, Darryl Holm and Igor Shevchenko.
 

We first consider multilevel Monte Carlo and stochastic collocation methods for determining statistical information about an output of interest that depends on the solution of a PDE with inputs that depend on random parameters. In our context, these methods connect a hierarchy of spatial grids to the amount of sampling done for a given grid, resulting in dramatic acceleration in the convergence of approximations. We then consider multifidelity methods for the same purpose which feature a variety of models that have different fidelities. For example, we could have coarser grid discretizations, reduced-order models, simplified physics, surrogates such as interpolants, and, in principle, even experimental data. No assumptions are made about the fidelity of the models relative to the “truth” model of interest so that unlike multilevel methods, there is no a priori model hierarchy available. However, our approach can still greatly accelerate the convergence of approximations.

In the past few decades, power grids across the world have become dependent on markets that aim to efficiently match supply with demand at all times via a variety of pricing and auction mechanisms. These markets are based on models that capture interactions between producers, transmission and consumers. Energy producers typically maximize profits by optimally allocating and scheduling resources over time. A dynamic equilibrium aims to determine prices and dispatches that can be transmitted over the electricity grid to satisfy evolving consumer requirements for energy at different locations and times. Computation allows large scale practical implementations of socially optimal models to be solved as part of the market operation, and regulations can be imposed that aim to ensure competitive behaviour of market participants.

Questions remain that will be outlined in this presentation.

Firstly, the recent explosion in the use of renewable supply such as wind, solar and hydro has led to increased volatility in this system. We demonstrate how risk can impose significant costs on the system that are not modeled in the context of socially optimal power system markets and highlight the use of contracts to reduce or recover these costs. We also outline how battery storage can be used as an effective hedging instrument.

Secondly, how do we guarantee continued operation in rarely occuring situations and when failures occur and how do we price this robustness?

Thirdly, how do we guarantee appropriate participant behaviour? Specifically, is it possible for participants to develop strategies that move the system to operating points that are not socially optimal?

Fourthly, how do we ensure enough transmission (and generator) capacity in the long term, and how do we recover the costs of this enhanced infrastructure?
 

Advances in manufacturing technologies, most prominently in additive manufacturing or 3d printing, are making it possible to fabricate highly optimised products with increasing geometric and hierarchical complexity. This talk will introduce our ongoing work on design optimisation that combines CAD-compatible geometry representations, multiresolution geometry processing techniques and immersed finite elements with classical shape and topology calculus. As example applications,the shape optimisation of mechanical structures and electromechanical components, and the topology optimisation of lattice-skin structures will be discussed.

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).

In this talk we will introduce and analyse a class of robust numerical methods for nonlocal possibly nonlinear diffusion and convection-diffusion equations. Diffusion and convection-diffusion models are popular in Physics, Chemistry, Engineering, and Economics, and in many models the diffusion is anomalous or nonlocal. This means that the underlying “particle" distributions are not Gaussian, but rather follow more general Levy distributions, distributions that need not have second moments and can satisfy (generalised) central limit theorems. We will focus on models with nonlinear possibly degenerate diffusions like fractional Porous Medium Equations, Fast Diffusion Equations, and Stefan (phase transition) Problems, with or without convection. The solutions of these problems can be very irregular and even possess shock discontinuities. The combination of nonlinear problems and irregular solutions makes these problems challenging to solve numerically.
The methods we will discuss are monotone finite difference quadrature methods that are robust in the sense that they “always” converge. By that we mean that under very weak assumptions, they converge to the correct generalised possibly discontinuous generalised solution. In some cases we can also obtain error estimates. The plan of the talk is: 1. to give a short introduction to the models, 2. explain the numerical methods, 3. give results and elements of the analysis for pure diffusion equations, and 4. give results and ideas of the analysis for convection-diffusion equations. 
 

A very common problem in Science is that we have some Data Observations and we are interested in either approximating the function underlying the data or computing some quantity of interest about this function.  This talk will discuss what are best algorithms for such tasks and how we can evaluate the performance of any such algorithm.