Start website main content

  • Istituto TeCIP - Telecomunicazioni, Informatica e Fotonica

"Dissemination On Secure AI for Edge Computing" partenariato esteso progetto PNRR SERICS

ai security
Date From 21.11.2025 time
End Date To 21.11.2025 time
Address

Via Alamanni 13 - zona artigianale La Fontina , San Giuliano Terme (PI) 56017 Italia

Back to Sant'Anna Magazine

Dissemination Workshop SERICS – Secure AI for Edge Computing
21 November 2025, 9:30 a.m.
TeCIP Institute, Scuola Superiore Sant’Anna – Grey Room

Scuola Superiore Sant’Anna is organizing a workshop dedicated to presenting the results achieved within the PNRR “Extended Partnership” PE7 SERICS – Security and Rights in the Cyber Space, Spoke 3, Science and Engineering Of Security of Artificial Intelligence (S.O.S. A.I.) project.

The workshop aims to share the progress made by the School’s research team within the SERICS project and to foster a discussion on new research directions to enhance the security of future embedded systems and AI applications for edge computing.

Agenda and programme attached.


ABSTRACT:

Recent advances in AI face challenges when applied to distributed embedded devices with specialized multimodal sensors, due to limited training data and computational resources. This scarcity makes models prone to overfitting, less reliable, and sensitive to environmental changes, while larger models cannot run efficiently. The work addresses these issues by leveraging self-supervised learning with unlabeled data, improving learning efficiency from scarce sensor data, and implementing scheduling and resource management to ensure timely and accurate inference. A case study on target classification and tracking using acoustic and seismic sensors demonstrates the benefits of these approaches on performance and responsiveness.


Brief bio of the speaker: 

Tarek Abdelzaher is a Full Professor of Computer Science at the University of Illinois at Urbana-Champaign, holding a Ph.D. from the University of Michigan (1999). He has authored over 450 publications in edge AI, IoT, real-time computing, sensor networks, and control, and has served as Editor-in-Chief of the Journal of Real-Time Systems. His research focuses on the performance and temporal properties of complex networked, embedded, and software systems. He has received multiple awards, including the IEEE Outstanding Technical Achievement and Leadership Award in Real-Time Systems and the Xerox Award for Faculty Research, and is a Fellow of IEEE and ACM.