Mathematics Webinar | AI Mathematics: Advanced Neural Networks Approximation
Mathematics Webinar | AI Mathematics: Advanced Neural Networks Approximation
Part of the MDPI Mathematics Webinars series
approximation by neural network operators, modulus of continuity, Jackson inequalities, Korovkin inequalities, quantitative results
Welcome from the Chair
22nd Mathematics Webinar
AI Mathematics: Advanced Neural Networks Approximation
In this webinar our two main features are as follows:
We will present the new idea of moving from the main neural network tools, the activation functions, to convolution integrals and singular integrals approximations. That is, the rare case of employing applied mathematics to treat theoretical ones.
We will introduce and also use the symmetrized neural network operators able to achieve supersonic speeds of convergence.
We will use a great variety of activation functions. Thus this webinar will present the original work by the speakers given at a very general level to cover a maximum number of different kinds of neural networks, covering ordinary, fractional and stochastic approximations. Univariate, fractional and multivariate approximations will also be presented. Iterated-sequential multi-layer approximations will be discussed as well.
Date: Friday, 10 April 2026
Time: 17:00 CEST to 18:45 CEST
Webinar ID: 845 9893 0293
Webinar Secretariat: journal.webinar@mdpi.com
Event Chairs
Department of Mathematical Sciences, University of Memphis, Memphis, TN 38152, USA
George A. Anastassiou was born in Athens, Greece in 1952. He received his B.SC degree in Mathematics from Athens University, Greece in 1975. He received his Diploma in Operations Research from Southampton University, UK in 1976. He also received his MA in Mathematics from University of Rochester, USA in 1981. He was awarded his Ph. D in Mathematics from University of Rochester, USA in 1984. During 1984-86 he served as a visiting assistant professor at the University of Rhode Island, USA. Since 1986 till 2024, he was a faculty member at the University of Memphis, USA. He has been a full Professor of Mathematics since 1994. From 2024 he is PROFESSOR EMERITUS. His research area is “Computational Analysis” in the very broad sense. He has published over 750 research articles in international mathematical journals and over 60-monographs, proceedings and textbooks in well-known publishing houses. Several awards have been awarded to George Anastassiou. In 2007 he received the Honorary Doctoral Degree from University of Oradea, Romania. He is associate editor in over 90 international mathematical journals and has been editor in-chief in 3 journals, most notably in the well-known “Journal of Computational Analysis and Applications”. G. ANASTASSIOU ON THE TOP 2% IN RESEARCHGATE.NET: https://www.researchgate.net/profile/George-Anastassiou/stats G. ANASTASSIOU ON THE WORLD’S TOP 2%: https://www.memphis.edu/mediaroom/releases/2023/december/uofm-research-top-2-percent-stanford-elsevier.php
Keynote Speakers
Department of Mathematical Sciences, University of Memphis, Memphis, TN 38152, USA
George A. Anastassiou was born in Athens, Greece, in 1952. He received his B.SC degree in Mathematics from Athens University, Greece, in 1975. He received his Diploma in Operations Research from Southampton University, UK, in 1976. He also received his MA in Mathematics from University of Rochester, USA, in 1981. He was awarded his PhD in Mathematics from University of Rochester, USA, in 1984. From 1984-86 he served as a visiting assistant professor at the University of Rhode Island, USA. From 1986 until 2024, he was a faculty member at the University of Memphis, USA. He has been a full Professor of Mathematics since 1994. Since 2024, he has been PROFESSOR EMERITUS. His research area is “Computational Analysis” in the very broad sense. He has published over 750 research articles in international mathematical journals and over 60 monographs, proceedings and textbooks in well-known publishing houses. Several awards have been awarded to George Anastassiou. In 2007 he received the Honorary Doctoral Degree from University of Oradea, Romania. He is associate editor in over 90 international mathematical journals and has been editor in-chief in 3 journals, most notably in the well-known Journal of Computational Analysis and Applications. G. ANASTASSIOU ON THE TOP 2% IN RESEARCHGATE.NET: https://www.researchgate.net/profile/George-Anastassiou/stats G. ANASTASSIOU ON THE WORLD’S TOP 2%: https://www.memphis.edu/mediaroom/releases/2023/december/uofm-research-top-2-percent-stanford-elsevier.php
Department of Mathematics and Computer Science, University of Perugia, 06123 Perugia, Italy
Danilo Costarelli was born in Perugia (Italy), on September 29, 1986. He obtained his Degree in Mathematics in 2010 at the University of Perugia (supervisors: Prof. Gianluca Vinti and Prof. Domenico Candeloro). He achieved his Ph.D. in Mathematics (main topics: Approximation Theory and its Applications) in 2014 at the "Roma Tre University" of Roma (supervisor: Prof. Renato Spigler). Currently, he is Associate Professor at the Department of Mathematics and Computer Science of the University of Perugia. His main research topics are functional and real analysis, and approximation theory and its applications to signal and image processing.
Department of Mathematics and Computer Sciences, University of Perugia, Via Vanvitelli 1, Perugia, I-06123 Italy
Professor Bardaro won an academic competition for Full Professorship. Since June 2013, he has been Full Professor at the University of Perugia. He is a member of the Unione Matematica Italiana, American Mathematical Socielty, and GNAMPA of CNR. He is an editor of various international journals. His research activity concerns approximation theory in functional spaces and linear or nonlinear operators, including sampling series and integral operators. He also studied problems concerning fractional calculus. He attended various international meetings, including lectures and short comunications. He also organized some meetings and minisymposiums at international conferences.
Department of Software Engineering, Faculty of Engineering and Natural Sciences, Istanbul Atlas University, Kagithane, Istanbul 34408, Türkiye
Dr. Seda Karateke is an Associate Professor at Istanbul Atlas University. She is a top expert in AI and network approximation.
Webinar Recording
In this section, you will find the recording of this webinar to watch. Re-watch and share with your colleagues!
In this webinar, the speakers introduced more intrinsic and sophisticated quantitative approximation properties of Activated Neural Network Operators in their convergence to the unit operator. The employed sigmoid activation functions either led to a cusp composite activation function of compact support or to a multi-composite activation function of infinite domain. Numerical results were also presented that proved the superiority of this method as well as our other method of symmetrization. Moreover, the speakers reported important sampling theory results and their connections to Neural Networks. Last, but not least, the topic of Multivariate Neural Network Operators—including inverse theorems and convergence rate in the Lp-norm—was covered.
Four incredible key speakers led the event, which was well attended by over 60 people worldwide. We would like to thank everyone who was involved in making this webinar a success, as well as the organizing company MDPI and the host personnel.
Registration
This is a FREE webinar. After registering, you will receive a confirmation email containing information on how to join the webinar. Registrations with academic institutional email addresses will be prioritized.
Certificates of attendance will be delivered to those who attend the live webinar.
Can’t attend? Register anyway and we’ll let you know when the recording is available to watch.
Program
|
Speaker/Presentation |
Time in CEST |
|
Prof. George Anastassiou Chair Introduction |
17:00-17:05 |
|
Prof. George Anastassiou Composition of Activation functions and the Reduction to finite domain |
17:05-17:25 |
|
Assoc. Prof. Danilo Costarelli Multivariate Neural Network Operators: inverse theorems, convergence rate in the L^p-norm and by the dissimilarity index defined through the continuous SSIM |
17:25-17:45 |
|
Prof. Carlo Bardaro A notion of Mellin distance and its applications to error estimates |
17:45-18:05 |
|
Assoc. Prof. Seda Karateke Mathematics Meets AI: Half-Hyperbolic Activation Functions and Symmetric Neural Networks |
18:05-18:25 |
|
Q&A |
18:25-18:40 |
|
Closing of Webinar Prof. George Anastassiou |
18:40-18:45 |
Relevant Special Issues
"Application and Perspectives of Neural Networks"
Edited by Dr. Ali Mehrabi
Deadine for manuscript submissions: 30 August 2026
"New Advances in Neural Networks and Applications"
Edited by Prof. Dr. Xinwei Cao, Dr. Ameer Tamoor Khan, and Prof. Dr. Predrag S. Stanimirović
Deadine for manuscript submissions: 29 April 2026
Sponsors and Partners
Organizers
