Hansjörg Albrecher (Université de Lausanne)
Randomization and Dimensionality in Risk Modeling abstract
Abstract:
Over the years, randomization has proven to be a powerful tool in the modeling of risks on several levels: for computational purposes, in uncovering connections between different models, but also in the consideration and generation of physical and/or synthetic scenarios in risk management. A second, and in part connected, theme is the parsimonious and structure-preserving refinement of stochastic models via matrix-valued parameters, and related questions concerning the appropriate and effective dimension of models for a given purpose. This lecture will deal with various recent advances in these fields, and also illustrate concrete applications in insurance and finance, including the optimal design of reinsurance treaties and the probabilistic analysis of the profitability of blockchain mining.
10:00 • ETH Zentrum, Rämistrasse 101, Zürich, Building HG, Room G 43
Ivan Mitrofanov (ENS Paris)
Orders on metric spaces and asymptotic invariants abstract
Abstract:
We study order ratio function -- an asymptotic invariant that describes how well a given metric space can be ordered. We say that an order is "good" if it can be effectively (with sub-linear competitive ratio) used as an universal order for solving traveling salesman problem.We describe connections of order ratio function with more traditional invariants, such as hyperbolicity, Assouad-Nagata dimension, number of ends and doubling.The talk is based on joint works with Anna Erschler.
10:30 • Université de Genève, Conseil Général 7-9, Room 1-05
Bastian Rieck (TU München)
Topology-Driven Diffusion in Structured and Unstructured Data Sets abstract
Abstract:
Diffusion is a general umbrella term for information propagation schemes. In this talk, I will briefly summarise our recent work on leveraging such schemes on structured data sets, i.e. graphs, and unstructured ones, i.e. point clouds. I will demonstrate how topology-driven diffusion schemes can improve classification performance and help in creating new hierarchical summaries of complex data sets.
13:00 • EPF Lausanne, Salle tba
Jonathan Weinberger (TU Darmstadt)
Abstract:
As an alternative to set-theoretic foundations, homotopy type theory is a logical system which allows for reasoning about homotopical structures in an invariant and more intrinsic way.
Specifically, for the case of higher categories there exists an extended framework, due to Riehl-Shulman, to develop (∞,1)-category theory synthetically. The idea is to work internally to simplicial spaces, where one can define predicates witnessing that a type is (complete) Segal. This had also independently been suggested by Joyal.
Generalizing Riehl-Shulman’s previous work on synthetic discrete fibrations, we discuss the case of synthetic cartesian fibrations in this setting, leading up to a 2-Yoneda Lemma. In developing this theory, we are led by Riehl–Verity’s model-independent higher category theory, therefore adapting results from ∞-cosmos theory to the type-theoretic setting. Time permits, we’ll briefly point out generalizations to the two-sided case.
In fact, by Shulman’s recent work on strict universes, the theory at hand has semantics in Reedy fibrant diagrams in an arbitrary (∞,1)-topos, so all type-theoretically formulated results semantically translate to statements about internal (∞,1)-categories.
This is based on joint work with Ulrik Buchholtz(https://arxiv.org/abs/2105.01724) and the speaker’s recent PhD thesis.
14:15 • EPF Lausanne, room MA A1 12 and online
Donald Youmans (Universität Bern)
Drinfeld-Sokolov reduction of 2d BF theory and the Schwarzian theory abstract
Abstract:
In this talk we report on the holographic correspondence between 2d BF theory on the (punctured) disk with gauge group PSL(2,R) and the Schwarzian theory on the boundary. We realize the duality in terms of a Drinfeld-Sokolov reduction which is equivalent to enforcing a first class constraint. The constrained path integral localizes over (exceptional) Virasoro coadjoint orbits and can be computed by a formal Duistermaat-Heckman integration.
The talk is based on joint work with F. Valach.
15:30 • Université de Genève, Conseil Général 7-9, Room 1-07 and zoom
Dr. Babak Maboudi Afkham (Delhi Technological University, India)
Abstract:
In my talk, I will describe a Bayesian framework for the X-ray computed tomography (CT) problem in an infinite-dimensional setting. We consider reconstructing piecewise smooth fields with discontinuities where the interface between regions is not known. Furthermore, we quantify the uncertainty in the prediction. Directly detecting the discontinuities, instead of reconstructing the entire image, drastically reduces the dimension of the problem. Therefore, the posterior distribution can be approximated with a relatively small number of samples. We show that our method provides an excellent platform for challenging X-ray CT scenarios (e.g. in case of noisy data, limited angle, or sparse angle imaging). We investigate the accuracy and the efficiency of our method on synthetic data. Furthermore, we apply the method to the real-world data, tomographic X-ray data of a lotus root filled with attenuating objects. The numerical results indicate that our method provides an accurate method in detecting boundaries between piecewise smooth regions and quantifies the uncertainty in the prediction, in the context of X-ray CT.
16:15 • EPF Lausanne, online seminar
Giada Franz (ETHZ)
What is... a minimal surface? abstract
Abstract:
Minimal surfaces are very natural objects to consider since they are surfaces that minimize the area, globally or just locally. <BR>This talk is meant to be a general introduction to these objects in different settings and an overview of some interesting questions and results, mainly related to the problem of the existence of minimal surfaces.
16:30 • UZH Zentrum, Building KO2, Room F 150
Dr. Joao Pereira (UT Austin)
Landscape Analysis of an improved power method for tensor decomposition abstract
Abstract:
In this talk, I will talk about the optimization formulation for symmetric tensor decomposition recently introduced in the Subspace Power Method (SPM) by me and Joe Kileel. Unlike popular alternative functionals for tensor decomposition, the SPM objective function has the desirable properties that its maximal value is known in advance, and its global optima are exactly the rank-1 components of the tensor when the input is sufficiently low-rank. We derive quantitative bounds such that any second-order critical point with SPM objective value exceeding the bound must equal a tensor component in the noiseless case and must approximate a tensor component if the tensor is approximately low-rank. We obtain a near-global guarantee for an overcomplete random tensor model, and a global guarantee assuming deterministic frame conditions. This implies that SPM with suitable initialization is a provable, efficient, robust algorithm for low-rank symmetric tensor decomposition. Time permitting, I will mention work in progress on a version of SPM for implicit tensor decomposition, that decomposes moment tensors without explicitly calculating these, and is much faster for this type of tensors.