Vladimir Fomichov (Aarhus University)
Abstract:
In our talk, we will consider random point processes formed by the clusters in one- dimensional coalescing particle systems and connected to the real Ginibre ensemble. Basing on recent work of R. Tribe and O. Zaboronski, we will first discuss their Pfaffian structure in the Brownian case, and then, by proposing a new algebraic proof, prove their conjecture that the obtained results remain valid in the general case.Furthermore, we will speak about the concentration of measure in diffeomorphic approximations of Brownian coalescing particle systems. In particular, the high-level exceedances of the stationary processes formed by the corresponding densities will be studied. More precisely, using the Rice formula, we will compute the level-crossing intensity of these processes and establish its asymptotic behaviour as the height of the level tends to infinity.
09:30 • Universität Basel, Zoom Seminar
Mark Aguiar (Princeton University)
Abstract:
In this chapter, we use a benchmark limited-commitment model to explore key issues in the economics of sovereign debt. After highlighting conceptual issues that distinguish sovereign debt as well as reviewing a number of empirical facts, we use the model to discuss debt overhang, risk sharing, and capital flows in an environment of limited enforcement. We also discuss recent progress on default and renegotiation; self-fulfilling debt crises; and incomplete markets and their quantitative implications. We conclude with a brief assessment of the current state of the literature and highlight some directions for future research.
10:30 • EPF Lausanne, Quartier UNIL - Dorigny, Salle Extranef 126
Erik Burman (University College London)
Abstract:
In many applications both in medical science and in the geosciences the accurate approximation of solutions to wave equations is an important component for optimisation or inverse identification. Examples include Thermoacoustic imaging or high frequency ultrasound treatments in medicine (HIFU) or fault slip analysis in seismology. These problems have in common the need for computational solution of an inverse problem where the forward problem is set in a heterogeneous domain. Indeed typically the sound speed in the bulk domain jumps over material interfaces. Sometimes there is even a need for coupling of the acoustic and elastodynamic equations in the presence of liquid inclusions. In this talk we will give a snapshot of our ongoing work in these topics, motivated by two such applications: HIFU and the propagation of seismic waves. After a brief introduction of the applications we will first discuss the analysis of some approximation methods for inverse initial value problems subject to the wave equation. We will then consider a hybrid high order method for the approximation of wave propagation in heterogeneous media, using cut element techniques to avoid meshing of interfaces. Finally we will discuss some open problems that remain in order to understand the approximation of the inverse initial value problem in heterogeneous media using high order methods.
11:00 • Universität Basel, Zoom Seminar
Vanessa Jacquier (Universität Florenz)
Abstract:
We consider the Ising model on the hexagonal lattice evolving according to Metropolis dynamics. We study its metastable behavior in the limit of vanishing temperature when the system is immersed in a small external magnetic field. We determine the asymptotic properties of the transition time from the metastable to the stable state up to a multiplicative factor and study the relaxation time and the spectral gap of the Markov process. We give a geometrical description of the critical configurations and show how not only their size but their shape varies depending on the thermodynamical parameters. Finally we provide some results concerning polyiamonds of maximal area and minimal perimeter.
14:00 • Universität Basel, Zoom Seminar
Björn Sprungk (TU Bergakademie Freiberg, Deutschland)
Laplace approximation and robust Monte Carlo for Bayesian inference with infomative data
16:00 • Universität Bern, Alpeneggstrasse 22, 3012 Bern, Hörsaal -203
Prof. Dr. Ben Bakker (University of Illinois at Chicago)
Algebraic approximation of Calabi-Yau varieties and the decomposition theorem abstract
Abstract:
Calabi-Yau manifolds are built out of simple pieces by the Beauville–Bogomolov decomposition theorem: up to an etale cover, any Kahler manifold with torsion first Chern class is a product of complex tori, irreducible holomorphic symplectic manifolds, and strict Calabi-Yau manifolds (which have no holomorphic forms except a holomorphic volume form). Work of Druel-Guenancia-Greb-Horing-Kebekus-Peternell over the last decade has culminated in a generalization of this result to projective Calabi-Yau varieties with the kinds of singularities that arise in the MMP, though the proofs critically use algebraic methods. In this talk I will describe joint work with H. Guenancia and C. Lehn on the deformation theory of Calabi-Yau varieties. Among other things, this allows us to extend the decomposition theorem to Kahler varieties and also resolves the K-trivial case of a conjecture of Peternell asserting that any minimal Kahler variety can be approximated by algebraic varieties.
16:00 • Zoom
Cynthia Rudin (Duke University, South Carolina)
Abstract:
With widespread use of machine learning, there have been serious societal consequences from using black box models for high-stakes decisions, including flawed bail and parole decisions in criminal justice, flawed models in healthcare, and black box loan decisions in finance. Transparency and interpretability of machine learning models is critical in high stakes decisions. In this talk, I will focus on two of the most fundamental and important problems in the field of interpretable machine learning: optimal sparse decision trees and optimal scoring systems. I will also briefly describe work on interpretable neural networks for computer vision.
Optimal sparse decision trees: We want to find trees that maximize accuracy and minimize the number of leaves in the tree (sparsity). This is an NP hard optimization problem with no polynomial time approximation. I will present the first practical algorithm for solving this problem, which uses a highly customized dynamic-programming-with-bounds procedure, computational reuse, specialized data structures, analytical bounds, and bit-vector computations.
Optimal scoring systems: Scoring systems are sparse linear models with integer coefficients. Traditionally, scoring systems have been designed using manual feature elimination on logistic regression models, with a post-processing step where coefficients have been rounded. However, this process can fail badly to produce optimal (or near optimal) solutions. I will present a novel cutting plane method for producing scoring systems from data. The solutions are globally optimal according to the logistic loss, regularized by the number of terms (sparsity), with coefficients constrained to be integers. Predictive models from our algorithm have been used for many medical and criminal justice applications, including in intensive care units in hospitals.
Interpretable neural networks for computer vision: We have developed a neural network that performs case-based reasoning. It aims to explains its reasoning process in a way that humans can understand, even for complex classification tasks such as bird identification.
Papers:
Jimmy Lin, Chudi Zhong, Diane Hu, Cynthia Rudin, Margo Seltzer
Generalized and Scalable Optimal Sparse Decision Trees. ICML, 2020.
Berk Ustun and Cynthia Rudin
Learning Optimized Risk Scores. JMLR, 2019. Shorter version at KDD 2017.
Struck et al.
Association of an Electroencephalography-Based Risk Score With Seizure Probability in Hospitalized Patients. JAMA Neurology, 2017.
Chaofan Chen, Oscar Li, Chaofan Tao, Alina Barnett, Jonathan Su, Cynthia Rudin
This Looks Like That: Deep Learning for Interpretable Image Recognition. NeurIPS, 2019.
16:15 • EPF Lausanne, Zoom Seminar
Dr. Inbar Seroussi (Weizmann Institute of Science, Israel)
How well can we generalize in high dimension? abstract
Abstract:
Deep learning algorithms operate in regimes that defy classical learning theory. Neural networks architectures often contain more parameters than training samples. Despite their huge complexity, the generalization error achieved on real data is small. In this talk, we aim to study generalization properties of algorithms in high dimension. Interestingly, we show that algorithms in high dimension require a small bias for good generalization. We show that this is indeed the case for deep neural networks in the overparametrized regime. In addition, we provide lower bounds on the generalization error in various settings for any algorithm. We calculate such bounds using random matrix theory (RMT). We will review the connection between deep neural network and RMT and existing results. These bounds are particularly useful when the analytic evaluation of standard performance bounds is not possible due to the complexity and nonlinearity of the model. The bounds can serve as a benchmark for testing performance and optimizing the design of actual learning algorithms. (Joint work with Prof. Ofer Zeitouni)