November 2018 - Random - Mailing list — Dipartimento di Matematica, UniPI

Concorso da RTDb di MAT/06 a Padova - scadenza 25/8/2016
by Marco Ferrante 25 Jun '24

25 Jun '24

Cari colleghi, vi segnalo che il Dipartimento di Matematica dell'Università di Padova ha bandito un posto di ricercatore di tipo B nel settore MAT/06. Il bando è disponibile alla pagina http://www.math.unipd.it/it/news/?id=1898 La scadenza per la presentazione delle domande è il 25 agosto 2016. Marco Ferrante

1 1

Seminario KOLOKOLTSOV @ UniTo
by Cristina Zucca 14 Oct '19

14 Oct '19

********************************************************************* SEMINARI DI PROBABILITA' E STATISTICA MATEMATICA DIPARTIMENTO DI MATEMATICA "G. PEANO" UNIVERSITA' DEGLI STUDI DI TORINO ********************************************************************* Martedì 17 Aprile 2018 alle ore 15:00 in Aula Marro presso il Dipartimento di Matematica "G. Peano" dell'Università degli Studi di Torino, Via Carlo Alberto 10, Il Prof. *VASSILI KOLOKOLTSOV* terrà un seminario dal titolo *Recent developments in probabilistic and analytic methods for fractional differential equations * Abstract: We shall discuss recent achievements in solving fractional differential equations and their extensions arising from generalized fractional calculus. Probabilistic techniques will be derived from the notion of Markov processes interrupted on the attempt to cross the boundary. Analytic techniques will cover generalized operator-valued Mittag-Leffler functions and chronological Feynman-Kac formulas. Tutti gli interessati sono invitati a partecipare. -- Dr. Cristina Zucca Department of Mathematics University of Torino Via Carlo Alberto 10 10123 Torino, Italy tel: +39-011-6702850 fax: +39-011-6702878

1 1

Summer school + Workshop on Chemical Reaction Networks Torino 24 June - 3 July 2018
by Enrico Bibbona 17 Jan '19

17 Jan '19

Dear colleagues, I'm organising a summer school and a Workshop on Chemical Reaction Networks in Torino, Italy, form June 24 to July 3rd 2019. I kindly ask you if you are interested in participating and to advertise those events to all the people you know that could be interested, in particular to the youngest. The events are described at the following link https://areeweb.polito.it/disma-excellence/CRNs/Index.html and a summary of the main informations follows. School (June 24-28 2019): It is intended for Master students, PhD and post Doc in Mathematics, Physics, Computer Science, Biology, Statistics, Engineering... More senior attendants are welcome as well. Two courses are organised, the instructors are Carsten Wiuf (Copenhagen University) and Greg Rempala (Ohio State University), and they will cover the theory of CRNs (both stochastic and deterministic) and some applications to molecular biology and epidemiology, including statistical inference. The school will be held in Pracatinat, a tiny mountain hamlet in the Chisone Valley not far from Torino. There is no registration fee, local expenses for a limited amount of early booking participants are covered by the organisation (these participants need to cover only their travel expenses). Workshop (July 1-3 2019) It is located at Politecnico di Torino, and will adress both stochastic and deterministic CRNs, modelling and inference. There is no registration fee, but participants are asked to cover their own expenses (we have limited funding for a few motivated exceptions). Whoever is interested in presenting a talk is welcome to preregister at the conference website. Best regards, Enrico Bibbona Probability, Statistics and Optimization group Department of Mathematical Sciences "G. L Lagrange" Politecnico di Torino

1 1

Postdoc position in Geometric Methods for Machine Learning and Deep Learning at RIST
by Luigi Malagò 14 Dec '18

14 Dec '18

=========================================================== *Subject*: Postdoc position in Machine Learning (1 year, renewable up to 2 years) *Institution*: RIST - Romanian Institute of Science and Technology, Cluj-Napoca *Keywords*: Geometry of Latent Spaces, Optimization over Manifolds, Information Geometry, Natural Gradient, Riemannian Geometry *Application deadline*: 10 October 2018 (applicants are encouraged to apply earlier) *Salary*: around 2200 euro net *Official announcement*: http://rist.ro/en/details/news/postdoc-positions-in-deep-learning-and-machi… =========================================================== Dear colleagues, the Romanian Institute of Science and Technology (RIST) has an opening for a postdoc position, in the context of the DeepRiemann project “Riemannian Optimization Methods for Deep Learning”, funded by European structural funds through the Competitiveness Operational Program (POC 2014-2020). The appointments will be for 1 year, with possible extensions up to 2 years. The DeepRiemann project aims at the design and analysis of novel training algorithms for Neural Networks in Deep Learning, by applying notions of Riemannian optimization and differential geometry. The task of the training a Neural Network is studied by employing tools from Optimization over Manifolds and Information Geometry, by casting the learning process to an optimization problem defined over a statistical manifold, i.e., a set of probability distributions. The project is highly interdisciplinary, with competences spanning from Machine Learning to Optimization, Deep Learning, Statistics, and Differential Geometry. The objectives of the project are multiple and include both theoretical and applied research, together with industrial activities oriented to transfer knowledge, from the institute to a startup or spin-off of the research group. The positions will be part of the new Machine Learning and Optimization group https://rist.ro/en/teams.html, which performs research at the intersection of Machine Learning, Stochastic Optimization, Deep Learning, and Optimization over Manifolds, using geometric methods based on Information Geometry. The group is one of two newly-formed groups in Machine Learning at RIST. The official job announcement can be seen here: http://rist.ro/en/details/news/postdoc-positions-in-deep-learning-and-machi… Informal inquiries can be sent to Dr. Luigi Malagò <malago(a)rist.ro>, principal investigator of the DeepRiemann project. Application deadline: 10 October 2018 (applicants are encouraged to apply earlier) best regards, Luigi Malagò

1 1

Final Announcement QFW2019 ETH Zurich
by Maria Elvira Mancino 03 Dec '18

03 Dec '18

Date: Fri, 13 Jul 2018 18:56:30 +0200 From: Josef Teichmann <jteichma(a)math.ethz.ch> CC: christa.cuchiero(a)univie.ac.at, qfw2019(a)math.ethz.ch Subject: 1st announcement QFW2019 Zurich User-Agent: Mutt/1.9.1 (2017-09-22) Dear colleagues, it is our great pleasure to announce the XX edition of the Workshop on Quantitative Finance (QFW2019) held at ETH Zurich The website of QFW2019 is https://people.math.ethz.ch/~jteichma/index.php?content=qfw2019 and you can find all relevant information there. The conference takes place from January 23 to January 25, 2019 and will last for three full days. You can just participate (registration is necessary and can be done by simply replying to this email providing "name + dates of attendance + intention to participate at the conference dinner") but you are warmheartedly invited to submit a paper: deadline for the submission of an extended abstract is November 30, deadline for submitting the paper is December 28, 2018. Accepted papers will be presented and discussed at the conference. With our best regards. Christa Cuchiero and Josef Teichmann -- Prof Josef Teichmann Department of Mathematics ETH Zürich Rämistrasse 101 8092 Zürich Switzerland fon +41 44 632 31 74 mobil +41 79 584 55 40 email jteichma(a)math.ethz.ch web www.math.ethz.ch/~jteichma

1 1

2 years postdoc position in statistics
by Guillaume Kon Kam King 30 Nov '18

30 Nov '18

Dear all, Benoit Liquet (https://sites.google.com/site/benoitliquet/home) is advertising a postdoc position in Statistics, located in the South-West of France, near Biarritz. More details below, and in attachment. Please forward if you know any interested candidate. Best, Guillaume KON KAM KING Università degli Studi di Torino *https://sites.google.com/site/guillaumekonkamking/home <https://sites.google.com/site/guillaumekonkamking/home>* Recruitment grade: young researcher (i.e. with PhD) Location: Anglet, France Duration: 12 to 24 months, starting late 2018 (December or January 2019) Deadline: position open until filled Gross Salary Range: 2699 euros / month Postdoc description: Recently, several approaches were developed for completing the agnostic > genome-wide association studies (GWAS) in the discovery of new genetic risk > factors. One such approach is gene set analysis (GSA) that incorporates the > available biological knowledge of genes in order to provide additional > insights into the mechanisms involved in carcinogenesis. Another approach > proposed to test variants for association for multiple phenotypes as over > the years multiple loci have been found to be associated to several > distinct traits revealing the existence of pleiotropic effects. Although a > number of statistical approaches have been proposed in the literature for > identifying the existence of pleiotropy at SNP-level, not a lot of work has > been conducted for investigating pleiotropy at gene- and pathway-levels. We > aim to fill this gap by developing advanced statistical methods that test > for pleiotropic effects at a pathway level. Also, we will extend these > methods for the analysis of GxE interactions, as to the best of our > knowledge, pleiotropy methods have not adapted for testing for GxE > interactions. Therefore, our first objective is to develop novel > statistical methods that will identify shared pathways between different > diseases. Our second aim is to extend these methods for testing GxE in > general or in the context of pleiotropy. These methods will be applied to > explore the genetic relationship between differentiated thyroid cancer > (DTC) and breast cancer (BC) types. It has been shown that these two > cancers are linked as women with a prior history of DTC are at an increased > risk for BC and conversely. Both tumors have been shown to be associated to > common hormonal risk factors and shared genetic susceptibility is > suspected. We will use data from the CECILE study (1019 BC cases and 999 > controls) and from EPITHYR (1345 DTC cases and 1399 controls). All subjects > were genotyped using the OncoArray chip that includes more than 500 000 > variants. Through the developments of novel statistical methods, we aim to > elucidate the common mechanisms between BC and DTC. This project will lead > to the development of new statistical methods that will benefit both the > epidemiology and statistical communities. These methods will permit to gain > novel insights into the inherited genetic basis of BC and DTC by > identifying the common biological elements that contribute to their > aetiology.

1 0

2 years postdoc position in statistics
by Guillaume Kon Kam King 30 Nov '18

30 Nov '18

Dear all, Benoit Liquet (https://sites.google.com/site/benoitliquet/home) is advertising a postdoc position in Statistics, located in the South-West of France, near Biarritz. More details below, and in attachment. Please forward if you know any interested candidate. Best, Guillaume KON KAM KING Università degli Studi di Torino *https://sites.google.com/site/guillaumekonkamking/home <https://sites.google.com/site/guillaumekonkamking/home>* Recruitment grade: young researcher (i.e. with PhD) Location: Anglet, France Duration: 12 to 24 months, starting late 2018 (December or January 2019) Deadline: position open until filled Gross Salary Range: 2699 euros / month Postdoc description: Recently, several approaches were developed for completing the agnostic > genome-wide association studies (GWAS) in the discovery of new genetic risk > factors. One such approach is gene set analysis (GSA) that incorporates the > available biological knowledge of genes in order to provide additional > insights into the mechanisms involved in carcinogenesis. Another approach > proposed to test variants for association for multiple phenotypes as over > the years multiple loci have been found to be associated to several > distinct traits revealing the existence of pleiotropic effects. Although a > number of statistical approaches have been proposed in the literature for > identifying the existence of pleiotropy at SNP-level, not a lot of work has > been conducted for investigating pleiotropy at gene- and pathway-levels. We > aim to fill this gap by developing advanced statistical methods that test > for pleiotropic effects at a pathway level. Also, we will extend these > methods for the analysis of GxE interactions, as to the best of our > knowledge, pleiotropy methods have not adapted for testing for GxE > interactions. Therefore, our first objective is to develop novel > statistical methods that will identify shared pathways between different > diseases. Our second aim is to extend these methods for testing GxE in > general or in the context of pleiotropy. These methods will be applied to > explore the genetic relationship between differentiated thyroid cancer > (DTC) and breast cancer (BC) types. It has been shown that these two > cancers are linked as women with a prior history of DTC are at an increased > risk for BC and conversely. Both tumors have been shown to be associated to > common hormonal risk factors and shared genetic susceptibility is > suspected. We will use data from the CECILE study (1019 BC cases and 999 > controls) and from EPITHYR (1345 DTC cases and 1399 controls). All subjects > were genotyped using the OncoArray chip that includes more than 500 000 > variants. Through the developments of novel statistical methods, we aim to > elucidate the common mechanisms between BC and DTC. This project will lead > to the development of new statistical methods that will benefit both the > epidemiology and statistical communities. These methods will permit to gain > novel insights into the inherited genetic basis of BC and DTC by > identifying the common biological elements that contribute to their > aetiology.

1 0

Avviso di seminario :: Causeur@DSS (Scienze Statistiche, Sapienza)
by Pierpaolo Brutti 30 Nov '18

30 Nov '18

----------------------------------------------------------------------------- A v v i s o d i S e m i n a r i o ----------------------------------------------------------------------------- Mercoledì 05 Dicembre, ore 11:00am ----------------------------------------------------------------------------- Sala 34 (quarto piano) Dipartimento di Scienze Statistiche Sapienza Università di Roma DAVID CAUSEUR (Agrocampus Ouest, Dep. of Statistics and Computer Science) terrà un seminario dal titolo HANDLING DEPENDENCE OR NOT IN STATISTICAL LEARNING FOR HIGH-DIMENSIONAL DATA tutti gli interessati sono invitati a partecipare. ----------------------------------------------------------------------------- Saluti Pierpaolo Brutti --- ABSTRACT The proper way to handle dependence across features in high-throughput data has raised fundamental discussions with unclear general conclusions or final recommendations. One of the most obvious illustration of this point is the tremendous effort of the statistics research community to address the impact of dependence on the False Discovery Rate (FDR)-controlling method by Benjamini and Hochberg (1995), which was initially designed under an independence assumption. Another famous questioning example is provided by the strikingly good performance of a naïve Bayes procedure ignoring dependence in a comparative study of machine learning methods by Dudoit et al. (2002) to predict classes from gene expression data. Addressing the dependence issue has often consisted in assessing its detrimental impact on the performance of standard methods designed to be optimal under independence, and deduce patches. To be valid for arbitrarily complex dependence patterns, such approaches in which dependence is viewed as a curse can lead to poorly powerful procedures. Therefore, both for machine learning and testing issues, a new generation of methods have emerged, advocating for an ad-hoc handling of dependence consisting in a preliminary whitening of the data (see Ahdesmäki and Strimmer, 2010, Hall and Jin, 2010). However, disentangling the dependent noise and the true association signal is very challenging and decorrelation can then lead to an alteration of the true association signal. For the purpose of global testing, where the objective is to test for the significance of an association signal between a set of features and a covariate, Arias-Castro el al. (2011) suggests that the optimal handling of dependence shall be specific of the pattern of the true association signal, especially through its sparsity rate. The former global testing framework covers a wide scope of applications, such as functional Analysis of Variance (fANOVA) and association tests between a region of the genome formed by contiguous Single Nucleotide Polymorphisms (SNP) and a case/control response variable in Genome Wide Association Studies. Interestingly, in the two former fields of applications, many popular methods are just based on simple aggregation of pointwise test statistics ignoring their dependence. The talk will start by a selection of short stories with confusing conclusions about the proper way to handle dependence. After a tentative clarification, I will introduce two methods for global testing in which whitening is adapted both to the pattern of dependence across pointwise test statistics and to the pattern of the true signal. The performance of the two testing methods will be illustrated by applications to significance analysis of ElectroEncephalogram curves in Event-Related Potentials (ERP) designs and by SNPset approaches of Genome Wide Association Studies for genetic epidemiology issues. We also discuss the applications of the former general principles to prediction in high-dimension.

1 0

Conference in honor of Aad van der Vaart
by Antonio Lijoi 30 Nov '18

30 Nov '18

Cari tutti, su gentile richiesta degli organizzatori, inoltro l'annuncio del convegno in onore di Aad van der Vaart. Saluti, AL ------------ Dear Colleagues, We are organising a conference between the 17th and 21st of June, 2019, in Leiden celebrating Aad van der Vaart's contributions to statistics on the occasion of his 60th birthday.If you are interested in participating, please register as soon as possible on the conference webpage (we have only limited space at the conference venue and registration is on first come first served base): http://pub.math.leidenuniv.nl/~schmidthieberaj/aadbirthday/index.html (if the link does not work, please try to copy it to the web browser.) The conference fee is 200 euro, which includes the welcome reception on Monday afternoon, and sandwich lunch, coffee and tea during the conference. To complete your registration, please send the registration fee to the following account: Bank name: Rabobank account number: 0102468869 IBAN: NL78RABO 0102468869 Swiftcode: RABONL2U in the name of: Universiteit Leiden cs Fac. W&N *mention:* SAPnummer 2009008062 and your name We will regularly check the account and will send you a confirmation letter once we received the registration fee. For further information, please do not hesitate to contact us (via the email address: b.t.szabo(a)math.leidenuniv.nl) Best wishes and hope to see you next year in Leiden, Subhashis Ghoshal Mathisca de Gunst Jacqueline Meulman Stéphanie van der Pas Johannes Schmidt-Hieber Botond Szabo Harry van Zanten

1 0

2019 ABS Summer School on Bayesian Demography (Como, Italy)
by Consonni Guido 29 Nov '18

29 Nov '18

We are glad to announce that the 2019 Applied Bayesian Statistics Summer School (16th edition) will be held in the magnificent Villa del Grumello, Como (Italy) along the Lake Como shore. BAYESIAN DEMOGRAPHY 24-28 June, 2019 Lecturers: Adrian Raftery and Hana Ševcíková University of Washington, Seattle, USA Further details are provided below. Guido Consonni and Fabrizio Ruggeri ABS19 Directors Raffaele Argiento ABS19 Executive Director ---------------------------------------------------------------- ******************************* * ABS19 * ******************************* Applied Bayesian Statistics School BAYESIAN DEMOGRAPHY Villa del Grumello, Como, Italy 24-28 June, 2019 Lecturers: - Adrian Raftery, the Boeing International Professor of Statistics and Sociology - Hana Ševcíková, Senior Research Scientist Center for Statistics and the Social Sciences, Center for Studies in Demography and Ecology, University of Washington, Seattle, USA The School webpage is http://web.mi.imati.cnr.it/conferences/abs19/index.html Registration is now open. Please note that the conference room allows only for a very limited number of participants. COURSE OUTLINE Population projections have until recently usually been done deterministically. Recently, the United Nation Population Division adopted a probabilistic approach to project fertility, mortality and population for all countries. In this approach, the total fertility rate and female and male life expectancy at birth are projected using Bayesian hierarchical models estimated via Markov Chain Monte Carlo. They are then combined with a cohort component model, which yields probabilistic projection for any quantity of interest. The methodology is implemented in a suite of R packages, which has been used by the UN analysts to produce the most recent revision of the World Population Prospects. This course will teach the theory and practice of Bayesian probabilistic projections. Ideas of the Bayesian hierarchical modeling for the three main components of population change, fertility, mortality and migration, will be explained. In hands-on exercises, students will become familiar with the functionality of the R packages. PRACTICAL INFORMATION The school will replicate the successful format of the previous years, and will feature lectures and practical sessions, as well as participants' talks. It will start on Monday after lunch and end on Friday before lunch; Wednesday afternoon is free. Accommodation is available either at the Villa guesthouse or in downtown hotels (info will appear soon on the website). Como can be easily reached by train from Milan and its airports. More details are available on the website. Dipartimento di Scienze Statistiche Università Cattolica del Sacro Cuore Largo Gemelli, 1 20123 Milano; ITALY Ph: +39 02 7234 3049 https://sites.google.com/site/consonnibayes/home [http://Static.unicatt.it/layout/img/layout/5x1000.gif] Destina il tuo 5 per mille all’Università Cattolica CF 02133120150 www.unicatt.it/5permille<http://www.unicatt.it/5permille/>

1 1