The IEEE Computer Society Annual Symposium on VLSI was held in Cyprus on July 4-6, 2022.

The Symposium explores emerging trends and novel ideas and concepts covering a broad range of topics in the area of VLSI: from VLSI circuits, systems, and design methods, to system-level design issues, to bringing VLSI design to new areas and technologies such as nano- and molecular devices, security, artificial intelligence, and Internet-of-Things, etc. Among the emerging technologies, great attention was paid to in-memory and neuromorphic computing.

Among the keynote speakers, Elisabetta Farella from our partner FBK provided a presentation on “TinyML for tiny embedded systems” in which she referred also to the work done in MARVEL to develop audio and vision deep learning techniques to be run on board resource-constrained platforms for the IoT in smart cities. Bringing AI to the very edge through TinyML approaches enables responsible AI, which is efficient and sustainable but also preserves privacy by design. Sensitive data remains local and are immediately consumed (i.e. not stored), only events and non-sensitive data are sent. Or data are immediately anonymized.