Abstract
This talk presents a unified research agenda focused on developing scalable algorithms and systems that enhance the efficiency, accuracy, and practicality of modern computing—spanning hardware design automation, compressed AI models, and intelligent sensing. The first part of the presentation covers foundational contributions in Electronic Design Automation (EDA), including methodologies in the ASIC design flow, fast and accurate gate and interconnect delay estimation using current-source models, and techniques for efficient sparsification and solution of dense matrices in large-scale circuit simulation. These methods substantially reduce computational cost while preserving modeling fidelity. Building on these algorithmic principles, the talk then introduces advances in the compression of vision and language models, with emphasis on the Vanishing Contribution Technique, a structured pruning approach that reduces model size while maintaining predictive performance. The presentation continues with contributions to smart-city perception systems, addressing AI-driven traffic scene understanding using static LiDAR sensors and LiDAR/event-driven camera fusion for robust, low-latency environmental sensing in complex conditions. It concludes with an overview of additional interdisciplinary projects and a teaching philosophy that emphasizes hands-on problem-solving, strong algorithmic foundations, and cross-domain integration to prepare students for emerging challenges in modern hardware and AI.
About the Speaker
Charalampos Antoniadis received the Diploma degree (Hons.) in Computer and Communication Engineering, the M.S. degree in Computer Science, and the Ph.D. degree in Electrical and Computer Engineering from the University of Thessaly, Volos, Greece, in 2011, 2014, and 2019, respectively. He is currently a Research Scientist at the I2S (Integrated Intelligent Systems) Lab, King Abdullah University of Science and Technology (KAUST), Thuwal, Saudi Arabia. He previously served as a Postdoctoral Fellow at KAUST until 2025, affiliated with the Innovative Technologies Laboratory (ITL) and the I2S lab. Before joining KAUST, he spent one year as an Adjunct Lecturer at the University of Thessaly. His research interests include applied machine learning, numerical methods for simulation, and algorithmic and system-level optimization across both hardware and software domains. Dr. Antoniadis was a recipient of scholarships from the University of Thessaly for his M.Sc. studies and an HFRI (Hellenic Foundation for Research and Innovation) fellowship for his Ph.D. studies. He was also the lead member of the student team that achieved First Place in the ACM TAU 2020 and ACM TAU2 2021 international timing analysis contests.
