Abstract: Many current intelligence systems such as autonomous vehicles, security applications, automated diagnostic systems, and others, have digital image classification as a crucial part. Over time, extensive research efforts have introduced many classification techniques but convolutional neural networks (CNNs) brought a revolution in the field and have become a superior method that has drastically improved classification accuracy. Topics that have been the subject of research for years are nowadays one-hour classes for even undergraduates. However, new research topics were introduced along with convolutional neural networks. For example, the problem of finding the optimal CNN configuration and architecture is exponentially hard since there are numerous hyperparameters such as the number, type and order of layers, number of neurons in each layer, kernel size, optimization algorithm, padding, stride, and many others, that should be fine-tuned. Additionally, this tuning has to be done for each classification problem. For this exponentially hard problem often a rather crude method is used that guesses good starting values and estimates better values for the hyper-parameters (guestimating). Another method that is commonly used is grid search. Since this is an optimization problem, some recent studies tested different optimization metaheuristics such as swarm intelligence algorithms. The usage of swarm intelligence algorithms for finding CNNs’ configuration is extremely time-consuming but the improvement of the classification accuracy can be significant. This talk will cover recent advantages and challenges of CNN such as finding the optimal configuration problem of unknown features, and more.

Biography: Milan Tuba, Professor of Computer Science and Mathematics, Head of the Artificial Intelligence Project at Singidunum University and Vice-Rector of Research at Sinergia University, is included in both versions of the Stanford University list of 2% of the most influential scientists in the world in all disciplines, one for contribution during the entire career and other for contribution in the previous year (for years 2020, 2021 and 2022). He was Vice Rector for International Relations, Singidunum University, Belgrade, Head of the Department for Mathematical Sciences at State University of Novi Pazar and Dean of the Graduate School of Computer Science at John Naisbitt University. Prof. Tuba is the author or coauthor of around 300 scientific papers (cited more than 6,500 times, h-index 48) and editor, coeditor or member of the editorial board or scientific committee of number of scientific journals, Springer books, congresses and international conferences. He was invited and delivered more than 60 keynote and inaugural lectures at international conferences.
He received B. S. in Mathematics, M. S. in Mathematics, M. S. in Computer Science, M. Ph. in Computer Science, Ph.D. in Computer Science from University of Belgrade and New York University. From 1983 to 1994 he was in the U.S.A. at Vanderbilt University in Nashville and Courant Institute of Mathematical Sciences, New York University and later as Assistant Professor of Electrical Engineering at Cooper Union School of Engineering, New York. During that time, he was the founder and director of Microprocessor Lab and VLSI Lab, leader of the NSF scientific projects and theses supervisor. He was the mentor of dozens of doctoral and master's dissertations at the Faculty of Mathematics University of Belgrade, Singidunum University, University of Sarajevo, State University of Novi Pazar, John Nesbitt University and University of East Sarajevo. He was teaching more than 20 graduate and undergraduate courses, from VLSI Design and Computer Architecture to Computer Networks, Operating Systems, Artificial Intelligence, Image Processing, Calculus and Queuing Theory at numerous universities in Europe and the USA. Prof. Tuba is a member of the National Agency for Accreditation of Universities of the Republic of Serbia.
His research interest includes Artificial Intelligence, Deep Learning, Neural Networks, Nature-inspired Optimization Algorithms, Image Processing, Computer Networks. Senior Member IEEE, ACM, AMS, SIAM, IFNA, Executive Board of IASEI.

More information about Prof. Milan Tuba could be found via https://eng.singidunum.ac.rs/profile/mtuba