Online webinar – February 25, 2022
Participants who will register by filling in this form will receive a link to participate live to the event, to ask questions to the speakers and participate to the discussion.
The webinar will be held on the platform Zoom, and will also be streamed on YouTube.
Nowadays the data deluge is the hallmark of a new kind of ‘Law of Large Numbers’ that builds intelligence from large, heterogeneous, noisy and, in general, complex data sets collected from mobile devices, the Internet of Things, software logs, automated medical devices, social media, and so many other data sources. The classical mathematical and statistical paradigms are often not applicable to current real-world problems, so there is a growing demand of new mathematical and statistical techniques to face such problems, able to shed some light on the often black-box techniques which are usually applied in such context and which characterise Deep Learning and, more generally, Artificial Intelligence. On the other side, since computers can not do everything by themselves, there is a growing need for new professional and scientific figures, the data scientists, that master a whole range of skills, ranging from data processing to sophisticated math tools and computational skills that are needed to extract the knowledge.
The state of the art and future research perspectives in this framework will be highlighted and discussed in this webinar by the PhD students enrolled in the EU funded MSCA Project BIGMATH (grant n. 812912), starting from a set of challenging industrial case studies.
Additionally some well known keynote speakers will introduce their point of view on possible future perspectives in different specific and quite hot subject areas related with the analysis of complex and big data.
PROGRAMME
All the times are referred to CET (Central European Time Zone)
9.00-9.15 Opening
9.15-9.30 Alessandra Micheletti, Università degli Studi di Milano – Overview of the BIGMATH Project (YouTube link)
The challenges of geometry
Chair: Alessandra Micheletti
9.30-10.15 Keynote lecture: Patrizio Frosini, Università degli Studi di Bologna – On the use of group equivariant non-expansive operators for topological data analysis and geometric deep learning. (YouTube link)
10.15-10.45 Filipa Valdeira, Università degli Studi di Milano – Gaussian Processes for shape modelling and registration. (YouTube link)
10.45-11.15 Rongjiao Ji, Università degli Studi di Milano – Group pattern detection of longitudinal data using a functional statistics framework. (YouTube link)
11.15-11.45 break
11.45-12.15 Stevo Rackovic, IST Lisbon – Increasing the Accuracy and Interpretability for Inverse Rig Solutions in the Human Face Animation (YouTube link)
Learning the data hidden lesson
Chair: M. Rosario Oliveira
12.15-13.00 Keynote lecture: Mário Figueiredo, Instituto de Telecomunicações, Instituto Superior Técnico, Universidade de Lisboa – Dealing with Strongly Correlated Variables in Supervised Learning (YouTube link)
13.00-14.30 lunch break
14.30 -15.00 Perfect Gidisu, TU-Eindhoven – Predicting customer investment potential value: An application in a Fintech company (YouTube link)
15.00-15.30 Rasool Taban, IST Lisbon – A novel balancing technique for mixed-type variables based on SMOTE (YouTube link)
Optimizing a complex world
Chair: Michiel Hochstenbach
15.30-16.15 Keynote lecture: Jose Mario Martinez, University of Campinas, Brazil –Block Coordinate Descent for smooth nonconvex constrained minimization (YouTube link)
16.15-16.45 break
16.45-17.15 Greta Malaspina, University of Novi Sad – A Modified Levenberg-Marquardt Method for Large Scale Network Adjustment (YouTube link)
17.15-17.45 Giulia Ferrandi, TU-Eindhoven – A harmonic framework for stepsize selection in gradient methods. An application to clustering problems (YouTube link)
17.45-18.00 closure
Organizing Committee
Michiel Hochstenbach, TU-Eindhoven
Natasa Krejic, University of Novi Sad
Alessandra Micheletti, Università degli Studi di Milano
Clàudia Nunes, IST Lisbon
M. Rosario Oliveira, IST Lisbon
Clàudia Soares, Nova University Lisbon
Contact: bigmath@unimi.it