January 2020 - Random - Mailing list — Dipartimento di Matematica, UniPI

Seminario at Scuola Normale Superiore, Tiziano De Angelis, School of Mathematics, The University of Leeds (January 22, 2020, 11:30)
by Giulia Livieri 17 Jan '20

17 Jan '20

Dear all, On Wednesday, 22 January 2020 — at 11:30 — at the Scuola Normale Superiore (Room: Aula Fermi) Dr Tiziano De Angelis (University of Leeds) will hold the following Seminar: Title Probabilistic results concerning smoothness of the value function and of the free boundary in optimal stopping Abstract I will present probabilistic proofs of some regularity properties for the value function of general optimal stopping problems and for the associated optimal boundaries. In particular this talk focuses on C^1 regularity of the value function and Lipschitz continuity of the optimal boundary. Most of our arguments rely on fundamental concepts from the theory of Markov processes and bridge the probabilistic and the analytical strands of the literature on free boundary problems. I will also illustrate situations in which our work improves or complements known facts from PDE theory. (based on https://epubs.siam.org/doi/pdf/10.1137/17M1113709 <https://epubs.siam.org/doi/pdf/10.1137/17M1113709> and https://arxiv.org/abs/1812.04564 <https://arxiv.org/abs/1812.04564>) Feel free to forward this announcement to your colleagues. Best Giulia Livieri

1 0

Seminar Cuturi @Bocconi
by Giacomo Zanella 16 Jan '20

16 Jan '20

We kindly invite you to the following BIDSA seminar: “Differentiable Ranking and Sorting using Optimal Transport” Marco Cuturi (Google Brain / ENSAE) Friday, January 17, 2020 | 12:30 | Room 3-e4-sr03 (third floor) Bocconi University, Via Roentgen 1, Milan Abstract: We propose in this paper proxy operators for ranking and sorting that are differentiable. To do so, we leverage the fact that sorting can be seen as a particular instance of the optimal transport (OT) problem between two univariate uniform probability measures, the first measure being supported on the family of input values of interest, the second supported on a family of increasing target values (e.g. 1,2,...,n if the input array has n elements). Building upon this link, we propose generalized rank and sort operators by considering arbitrary target measures supported on m values, where m can be smaller than n. We recover differentiable algorithms by adding to this generic OT problem an entropic regularization, and approximate its outputs using Sinkhorn iterations. To illustrate the versatility of these operators, we use the soft-rank operator to propose a new classification training loss that is a differentiable proxy of the 0/1 loss. Using the soft-sort operator, we propose a new differentiable loss for trimmed regression. Speaker: Marco Cuturi joined Google Brain, in Paris, in October 2018. He graduated from ENSAE (2001), ENS Cachan (Master MVA, 2002) and holds a PhD in applied maths obtained in 2005 at Ecole des Mines de Paris. He worked as a post-doctoral researcher at the Institute of Statistical Mathematics, Tokyo, between 11/2005 and 3/2007. He worked in the financial industry between 4/2007 and 9/2008. After working at the ORFE department of Princeton University between 02/2009 and 08/2010 as a lecturer, he was at the Graduate School of Informatics of Kyoto University between 9/2010 and 9/2016 as a tenured associate professor from 10/2013. He then joined ENSAE, the french national school for statistics and economics, in 9/2016, where he still teaches. His recent proposal to solve optimal transport using an entropic regularization has re-ignited interest in optimal transport and Wasserstein distances in the machine learning community. His work has recently focused on applying that loss function to problems involving probability distributions, e.g. topic models / dictionary learning for text and images, parametric inference for generative models, regression with a Wasserstein loss and probabilistic embeddings for words. Best regards, Giacomo Zanella Please note: if you are a guest and you do not have a Bocconi ID Card to access to the Bocconi Buildings, please communicate your participation by sending an email to c.monetti(a)unibocconi.it

1 1

Ernesto De Vito @ Roma Tor Vergata
by Anna Vidotto 16 Jan '20

16 Jan '20

Dear all, I would like to announce the following talk that will be held on Thursday January 23rd 2020, at 4:30 PM, in Aula D’Antoni of the Department of Mathematics, University of Rome Tor Vergata. Speaker: Ernesto De Vito Title: Multi-Scale Vector Quantization with Reconstruction Trees Abstract: The aim of vector quantization is a parsimonious reconstruction of unsupervised data as provided by K-means and geometric multi-resolution analysis, to name a few. In the talk I will present a multi-scale algorithm for vector quantization and study its statistical properties. The proposed approach is inspired by decision trees and is based a family of given partitions, to quickly explore the data in a coarse to fine-multi-scale-fashion. The main technical contribution is an analysis of the expected distortion achieved by the proposed algorithm. Both asymptotic and finite sample results are derived. The presentation is based on a joint work with E. Cecini and L. Rosasco. Kind regards --------- Anna Vidotto PostDoc Researcher Dipartimento di Matematica Università degli Studi di Roma Tor Vergata https://sites.google.com/view/annavidotto

1 0

seminario Djehiche a Milano 27/1/2020
by Marco Alessandro Fuhrman 15 Jan '20

15 Jan '20

Il prof. Boualem Djehiche (KTH Stockholm) terrà un seminario dal titolo On a class of mean-field reflected backward stochastic differential equations and the related obstacle problem. Il seminario è fissato alle ore 10:30 del 27/1/2020 presso il Dipartimento di Matematica dell'Università di Milano (via Saldini 50), in aula C (II piano). Tutti gli interessati sono cordialmente invitati. Un sommario è riportato sotto. Il prof. Djehiche è ospite del Dipartimento per tutta la settimana. Cordialmente, Marco Fuhrman --------------------------- Abstract. In this talk I will review some recent results on the solvability of a class of reflected backward stochastic differential equations (BSDEs) (driven by the Brownian motion) of mean-field type, where the mean-field interaction in terms of the distribution of the Y-component of the solution enters both the driver and the lower obstacle. When the reflected BSDE is driven by an underlying Markov diffusion, it is possible to derive a related obstacle problem in the form of a variational inequality involving differential operators w.r.t. both the space and the probability measure components. -------------------------- <signaturebeforequotedtext></signaturebeforequotedtext><signatureafterquotedtext></signatureafterquotedtext>

1 0

Post-doctoral position in probability - Paris
by Cristina TONINELLI 14 Jan '20

14 Jan '20

Dear friends and colleagues, I am currently advertising a post-doctoral position in Probability at Ceremade, University Paris Dauphine. The position is for two years. The expected starting date is September 2020, though a different date may be arranged. The position is funded by my European project MALIG, whose aim is to develop a mathematical approach to the liquid-glass transition and related phenomena. Main areas of research include interacting particle systems, bootstrap percolation cellular automata and glassy dynamics. The successful candidate will also have the opportunity to collaborate with the probability group of University Roma Tre, in particular with Fabio Martinelli, a key member of the MALIG team. The grant will provide substantial fundings to cover working periods in Rome. There are no teaching duties associated to the position. Candidates can contact me directly for any additional informations. Applications including a CV, a list of publications and a one-page description of research interests and experience should be sent directly to me at toninelli(a)ceremade.dauphine.fr<mailto:toninelli@ceremade.dauphine.fr> Applicants should also arrange for up to three letters of recommendation to be sent to the same address. The deadline for applications is Monday, 3rd February 2020. I would be grateful if you could bring this announcement to the attention of potential candidates and distribute it via local mailing lists. Best wishes, Cristina Toninelli --------------------------------------------------------------------------------------------- Cristina Toninelli Postal Address: Ceremade Univ.Paris Dauphine Place du Marechal de Lattre de Tassigny 75775 Paris Cedex 16 - France Office: B617 --------------------------------------------------------------------------------------------

1 0

de Finetti Risk Seminar - January 2020
by Emanuela Rosazza 14 Jan '20

14 Jan '20

On behalf of the Scientific Committee of the de Finetti Risk Seminars, we are glad to invite you to participate at the following Lecture Quadratic Hedging and Optimization of Option Exercise Policies in Incomplete Markets and Discrete Time Nicola Secomandi Tepper School of Business – Carnegie Mellon University Abstract. This paper extends quadratic hedging from European to Bermudan options in discrete time when markets are incomplete and investigates its use for supporting exercise policy optimization. The key idea is to construct date specific approximate replicating portfolios. Hedging any given exercise policy can be done by solving a collection of stochastic dynamic programs. Optimizing the exercise policy based on the resulting martingale measure requires care. If this measure is risk neutral (RN), the value of an optimal such policy, which can be obtained by augmenting the hedging model with an exercise policy optimization step, is a no arbitrage one. Otherwise this approach must be refined by imposing time consistency on exercise policies, although the value of the resulting exercise policy may not be arbitrage free. Following the common pragmatic strategy of specifying quadratic hedging under an RN measure, e.g., one calibrated to market prices, avoids these issues. In particular, it provides a simple hedging policy with immediate practical applicability and is equivalent to exercise policy optimization under RN valuation, thus complementing it with a consistent hedging policy. A simple numerical example shows that this procedure generates effective hedging policies LOCATION: The seminar will be held on Wednesday, January 15, at 18.00 at room 3-E4-SR03, Bocconi University, Via Rontgen 1, 3rd floor, Milano. Scientific Committee: Prof. Simone Cerreia-Voglio (Univ. Bocconi) Prof. Marco Frittelli (Univ. degli Studi di Milano) Prof. Fabio Maccheroni (Univ. Bocconi) Prof. Marco Maggis (Univ. degli Studi di Milano) Prof. Massimo Marinacci (Univ. Bocconi) Prof. Emanuela Rosazza Gianin (Univ. Milano-Bicocca) ******************************************************************************************* Emanuela Rosazza Gianin Dipartimento di Statistica e Metodi Quantitativi Università di Milano-Bicocca Edificio U7 – 4° Piano Via Bicocca degli Arcimboldi, 8 20126 Milano Tel. 02 64483208 Fax. 02 64483105 e-mail: emanuela.rosazza1(a)unimib.it *******************************************************************************************

1 0

Seminar Announcement: M. Glanzer (univie) at Polimi
by Carlo Sgarra 14 Jan '20

14 Jan '20

On Monday, January 21, at 11:00 o'clock, in the Seminar Room located at the third floor of the MATHEMATICS Department ( POLITECNICO di Milano) MARTIN GLANZER of the University of Vienna will give a talk on the following subject: Stochastic Optimization with Multiple Time Scales ( Joint work with G. Pflug) All interested people are kindly invited to participate. ABSTRACT Real-world multistage stochastic optimization problems are often characterized by the fact that the decision maker may take actions only at specific points in time, even if relevant data can be observed much more frequently. In such a case there are not only multiple decision stages present but also several observation periods between consecutive decisions, where profits/costs occur contingent on the stochastic evolution of some uncertainty factors. We present a tailor-made modeling framework for such problems, which allows for a computationally efficient solution. We first establish new results related to the approximation of (Markovian) stochastic processes by scenario lattices. In a second step, we incorporate the multiscale feature by leveraging the theory of stochastic bridge processes. The ingredients to our proposed modeling framework are elaborated explicitly for various popular examples, including both diffusion and jump models. In particular, we present new results related to the simulation of compound Poisson bridges. Finally, we discuss a valuation problem of a thermal power plant, where implementing our multiscale modeling framework turned out to be particularly convenient. If time permits, we incorporate model ambiguity into the power plant valuation problem and show some numerical results.

1 0

Seminar annoucement at POLIMI- Talk by Martin GLANZER (Univie)
by Carlo Sgarra 14 Jan '20

14 Jan '20

On Monday, January 21, in the Seminar Room located at the third floor of the MATHEMATICS Department ( POLITECNICO di Milano) MARTIN GLANZER of the University of Vienna will give a talk on the following subject: Stochastic Optimization with Multiple Time Scales ( Joint work with G. Pflug) All interested people are kindly invited to participate. ABSTRACT Real-world multistage stochastic optimization problems are often characterized by the fact that the decision maker may take actions only at specific points in time, even if relevant data can be observed much more frequently. In such a case there are not only multiple decision stages present but also several observation periods between consecutive decisions, where profits/costs occur contingent on the stochastic evolution of some uncertainty factors. We present a tailor-made modeling framework for such problems, which allows for a computationally efficient solution. We first establish new results related to the approximation of (Markovian) stochastic processes by scenario lattices. In a second step, we incorporate the multiscale feature by leveraging the theory of stochastic bridge processes. The ingredients to our proposed modeling framework are elaborated explicitly for various popular examples, including both diffusion and jump models. In particular, we present new results related to the simulation of compound Poisson bridges. Finally, we discuss a valuation problem of a thermal power plant, where implementing our multiscale modeling framework turned out to be particularly convenient. If time permits, we incorporate model ambiguity into the power plant valuation problem and show some numerical results.

1 0

Probability Seminar: Richard Kraaij (TUDelft) [Padova, Tue 14/01 (TOMORROW), 11am]
by Francesca Collet 13 Jan '20

13 Jan '20

*** REMINDER *** ____________________________________________________________________________________________________ DR. RICHARD C. KRAAIJ (Delft University of Technology) “Large deviations for weakly coupled slow-fast systems via the comparison principle of an associated Hamilton-Jacobi equation” Tuesday 14 January (TOMORROW), 11:00; room 1BC50, Dipartimento di Matematica “Tullio Levi-Civita”, Via Trieste 63, Padova ABSTRACT. In statistical physics many interesting phenomena, e.g. behavior of systems at critical parameters or in the theory of hydrodynamic limits, arise from systems having multiple time-scales. A slow component is influenced by fast components, and as the number of interacting components tends to infinity, limiting results for the slow component are obtained in terms of `averaged' versions of the fast components. I will consider in my talk the fluctuations (large deviations) of coupled Markovian systems with two-time scales. These fluctuations can arise from two sources: fluctuations of the slow process itself, or fluctuations of the large time averages of the fast process, effectively leading to a competition of two fluctuation effects. To obtain the large deviation principle, we consider an associated Hamilton-Jacobi-Bellman equation of which the Hamiltonian is given in terms of the two fluctuation effects. We establish under mild conditions that this Hamilton-Jacobi-Bellman equation is well-posed, and as a consequence that we have a large deviation principle for a wide class of weakly coupled Markov processes. Based on joint work with Mikola Schlottke (Eindhoven, The Netherlands). ____________________________________________________________________________________________________

1 0

OSD2020-14th International Conference on Ordered Statistical Data
by malongob＠unina.it 13 Jan '20

13 Jan '20

Dear all, we are glad to announce the 14th International Conference on Ordered Statistical Data - OSD2020, that will take place on May 26-29 at Vietri sul Mare (SA), Italy. The conference website is http://www.osd2020.unina.it/ The conference is devoted to recent advances and trends in the mathematical theory of ordered statistical data, including order statistics and their generalizations, and their applications in reliability, statistics and lifetime analysis. The scientific program will include plenary talks and sessions of contributed talks. It is our pleasure to point out the presence of Prof. Fabio Spizzichino in the list of keynote speakers of the conference. We look forward to welcome you in Vietri sul Mare. Chief organizers, Maria Longobardi and Franco Pellerey

1 0