Keynote Speakers

We are very pleased to have acquired the services of an excellent selection of keynote speakers for the symposium The speakers and the titles of their talks are shown below.

Prof. Anna Esposito

University of Campania "Luigi Vanvitelli", Italy

An overview on mood/emotion research: the effects of gender and age of stimuli
More details ...

Dr. Eng. Mihaela Luca

Institute of Computer Science, Romanian Academy, Romania

From Deep Learning to Vision Transformers - a big step further
More details ...

Prof. Gloria Phillips-Wren

Loyola University Maryland, USA

The Impact of Generative AI on Decision Making and Analytics
More details ...

Prof. Anna Esposito

University of Campania "Luigi Vanvitelli", Italy

An overview on mood/emotion research: the effects of gender and age of stimuli

Abstract:

In this talk I will present the current research performed at the Behaving Cognitive Systems (BeCogSys) laboratory at University of Campania "Luigi Vanvitelli" on human’s ability to decode emotional expressions and negative mood. Implications for their automatic detection and effects of gender and age of stimuli will be discussed.

Biography:

Anna Esposito received her “Laurea” Degree summa cum laude from Salerno University with a thesis on Neural Networks (Complex System, 6(6), 507-517, 1992). She received the PhD degree in Applied Mathematics and Computer Science from Naples University “Federico II”, with a PhD thesis developed at MIT, RLE Lab (Boston, USA) on mathematical models of speech production (Phonetica, 59(4), 197-231, 2002).

Anna has been postdoc at the International Institute for Advanced Scientific Studies, and Assistant Professor at Dept of Physics, Salerno University, where she has taught courses on Laboratory of Cybernetics, Neural Networks, and Speech Processing (1996-2000). From Nov 2000 to Nov 2002, she had a research professor position at Wright State University, Dept. of Computer Science and Engineering, OH, USA. Anna has been Associate professor (from 2003 to 2019) and currently is full professor in Cybernetics, Artificial Intelligence and Multimodal Communication at Università della Campania “Luigi Vanvitelli”, Department of Psychology, where she teaches course on Fundamentals of AI and Neural Networks, Cognitive and Algorithmic Issues of Multimodal Communication, Cognitive Economy and Decision Making. She is the head of the Behaving Cognitive Systems laboratory (BeCogSys, https://www.psicologia.unicampania.it/home-becogsy), the president of the Italian Society on Neural Networks (SIREN) https://www.siren-neural-net.it/, and the Chair of the IAPR Conferences & Meetings Committee, https://iapr.org/committees/conferences-and-meetings-committee.

During the last five years the lab participated/is participating to the H2020 projects: a) Empathic, www.empathic-project.eu/ and b) Menhir, menhir-project.eu/ , the Italian projects: c) SIROBOTICS, https://www.grifomultimedia.it/en/portfolio_page/si-robotics-social-robotics-for-active-and-healthy-ageing/, d) ANDROIDS, https://www.psicologia.unicampania.it/android-project, and SALICE: Socially-Aware Learning through Interactions in Crowded Environments, and the Erasmus project G-Guidance: https://g-guidance.eu/language/en/. Anna is member of the European Science Foundations (ESF) and the EU networks Cognition. She authored 300+ peer reviewed publications and edited/co-edited 32+ international books.

Dr. Eng. Mihaela Luca

Institute of Computer Science, Romanian Academy, Romania

From Deep Learning to Vision Transformers - a big step further

Abstract:

The transition from deep learning (DL) to Vision Transformers (ViTs) represents a significant leap and an important challenge in automatic image interpretation and analysis. This shift from traditional Deep Learning, starting with the CNN-based approaches which dominated medical image analysis due to their strong ability to capture spatial hierarchies in images, were used in disease detection (e.g., pneumonia detection in X-rays), tumor segmentation (e.g., brain tumors in MRI scans), histopathology image classification, organ segmentation in CT scans, polyps in colonoscopy, etc. As limitations, CNNs struggle with long-range dependencies in images, they rely heavily on local features and may require extensive data augmentation and their performance is sensitive to variations in imaging conditions. In the emergence of Vision Transformers (ViTs), a significant role played the Self-Attention Mechanism. Unlike CNNs, ViTs process images as a sequence of patch embeddings and use self-attention to capture both local and global dependencies. The move from deep learning (CNNs) to ViTs represents a paradigm shift in medical image processing, offering new creative capabilities while posing new challenges in interpretability, error analysis, and trustworthiness—which are critical for clinical adoption.

Biography:

Mihaela Luca (formerly Costin), https://orcid.org/0000-0001-8924-3161, is a scientific researcher in the Computer Science Institute, Romanian Academy, Iași branch. She obtained her PhD in 2005 with a thesis on the development of decision systems. As an Associate Professor, she taught disciplines in the area of Artificial Intelligence (Image Processing, Artificial Intelligence, Mechatronics and Expert Systems) to the students of the Computer Science Faculty of Alexandru Ioan Cuza University and from the Faculty of Automatic Control and Computer Engineering of Gheorghe Asachi Technical University of Iași, Romania. In 2023, she received the Excellence award of the Romanian Academy (ex-aequo) for her most recent accomplishments. Her research has focused mostly on two-dimensional signal processing, assessing images of interest in materials physics (nanostructures), medicine (scintigraphy, hematology, video colonoscopy and noninvasive cancer diagnosis using microwaves), employing artificial intelligence, neural networks, deep learning, fuzzy logic, and/or uncertainty reasoning. With over a hundred scientific papers and chapters in the last two decades, co-authoring three books and co-editing more volumes, she has been working in the frame of several scientific projects, either as coordinator for the Computer Science Institute in an excellence research national program for noninvasive breast cancer detection, or as a partner in an IAEA European Project using scintigraphy for cancer detection. She is a member in the editorial board of some international journals, reviewer for more interdisciplinary publications, and the scientific secretary of the Romanian Committee of History and Philosophy of Science and Techniques of the Romanian Academy, Iasi Branch, serving as the editorial secretary of Noesis Journal, belonging to the Romanian Academy, too. Her hobbies are oil painting and psychology.

Prof. Gloria Phillips-Wren

Loyola University Maryland, USA

The Impact of Generative AI on Decision Making and Analytics

Abstract:

Generative AI (genAI) has burst upon the workplace over the past few years impacting how work is accomplished and by whom. In particular, the type of genAI called Large Language Models (LLMs) have demonstrated the facility to provide analysis of data and what appears to be reasonable answers to questions posed and answered in natural language. The jury is still out on what tasks and whose jobs may be subsumed by genAI and LLMs. For example, in the area of decision making and analytics, creativity is often cited as a necessary part of analysis as the decision maker develops approaches to solve a decision problem. However, can genAI actually be creative when we know that it is a statistical parrot? In addition, should we trust analysis that we did not develop ourselves with human intelligence? In this discussion, we look at current research in genAI for decision making and go to the professional community for insight on how LLMs are being using to change analytics workflows. We conclude by suggesting future research questions related to decision making using this emerging technology.

Biography:

Gloria Phillips-Wren is a Professor in the Department of Information Systems, Law and Operations Management at Loyola University Maryland.  She is a founder and co-editor-in-chief of Intelligent Decision Technologies (IDT) and Associate Editor of the Journal of Decision Systems (JDS).  Dr. Phillips-Wren is a past-chair and Board Member of the Special Interest Group on Decision Support and Analytics (SIGDSA) under the Association of Information Systems (AIS), Secretary of IFIP WG8.3 Decision Support (DS), and leader of a focus group for KES in intelligent decision technologies.  Her research interests and publications are in decision making and support, analytics, generative AI, healthcare IT, and strategic uses of technologies such as genAI. Her work appears in Omega, Journal of Network and Computer Applications, Decision Support Systems, IDT, JDS, Journal of Organizational Computing and Electronic Commerce, Expert Systems with Applications, Big Data, IT & People, among others. She has also published 13 books (including co-edited), along with numerous book chapters and conference proceedings.