Edge AI (2.0) For Future Computing
Aaron Ding, TU Delft, Netherlands
Decentralised Machine Learning As an Enabler of Decentralised Online Services
Sonia Ben Mokhtar, LIRIS CNRS, France
Edge AI (2.0) For Future Computing
Aaron Ding
TU Delft
Netherlands
![]()
Brief Bio
Aaron Ding leads the Cyber Physical Intelligence (CPI) Lab as tenured Associate Professor at TU Delft. He has supervised 100+ students and won EU research grants (€5M+€5.5M) as Consortium Director & PI by leading a series of international and national R&D projects (EU Horizon, Marie Curie ITN, German DAAD, EIT Digital, Tekes Finland) in tight collaboration with industrial companies of Nokia, Ericsson, Deutsche Telekom, Broadcom, Telia, NEC, Telefónica, Fraunhofer and WithSecure. He has 17 years top R&D practices in EU, Switzerland, UK and USA. Prior to TU Delft, he has worked at TU Munich with Jörg Ott, at University of Cambridge with Jon Crowcroft, at Columbia University with Henning Schulzrinne. He had sabbatical in 2023 at ETH Zürich with Adrian Perrig. His PhD is on Computer Science with Sasu Tarkoma (Helsinki) and Jon Crowcroft (Cambridge). Funded by the Nokia Foundation, vital part of his PhD is completed at the University of Cambridge and Columbia University in New York. Being an active member of ACM, IEEE, Springer and IETF, he is the founder of ACM EdgeSys, Editor for ACM TIOT and Springer Nature Computing. He has served on chairing and program committees for prestigious international conferences including ACM SIGCOMM, ACM MobiCom, ACM UbiComp, ACM WWW, ACM CoNEXT, ACM MobiSys, ACM SenSys, ACM SEC, IEEE INFOCOM, IEEE ICDCS. He is a recipient of the esteemed Nokia Foundation Scholarships. His homepage at TU Delft: https://sk-alg.tbm.tudelft.nl/aaron/
Abstract
Similar to the progression from cloud computing to cloud intelligence, we are witnessing a fast evolution from edge computing to edge intelligence (aka Edge AI). As a rising research branch that merges distributed computing, data analytics, embedded and distributed ML, Edge AI is rapidly advancing and envisioned to provide adaptation for data-driven applications and enable the creation, optimization and deployment of distributed AI/ML pipelines. However, despite all the promises ahead, how to harness Edge AI for future computing is far from straightforward since critical building blocks are still missing. This keynote will illustrate major concerns behind Edge AI from two exemplary aspects, i.e., trustworthy and sustainable perspectives. Besides practical lessons from my latest EU Horizon SPATIAL and Marie Curie APROPOS projects on Edge AI, I will share an envisioned roadmap to shed light on the evolving path of Edge AI 2.0 for future computing.
Decentralised Machine Learning As an Enabler of Decentralised Online Services
Sonia Ben Mokhtar
LIRIS CNRS
France
Brief Bio
Sonia Ben Mokhtar is a CNRS research director at the LIRIS laboratory, Lyon, France and the head of the distributed systems and information retrieval group (DRIM). She received her PhD in 2007 from Université Pierre et Marie Curie before spending two years at University College London (UK). Her research focuses on the design of resilient and privacy-preserving distributed systems. Sonia has co-authored 80+ papers in peer-reviewed conferences and journals and has served on the editorial board of IEEE Transactions on Dependable and Secure Computing and co-chaired major conferences in the field of distributed systems (e.g., ACM Middleware, IEEE DSN). Sonia has served as chair of ACM SIGOPS France and as co-chair of GDR RSD a national academic network of researchers in distributed systems and networks.
Abstract
There is a strong momentum towards data-driven services at all layers of society and industry. This started from large scale web-based applications such as Web search engines (e.g., Google, Bing), social networks (e.g., Facebook, TikTok, Twitter, Instagram) and recommender systems (e.g., Amazon, Netflix) and is becoming increasingly pervasive thanks to the adoption of handheld devices and the advent of the Internet of Things. Recent initiatives such as Web 3.0 are coming with the promise of decentralising such services for empowering users with the ability to gain back control over their personal data, and prevent a few economic actors from over concentrating decision power. However decentralising online services calls for decentralising the data and the machine learning algorithms on which they heavily rely. While Federated Learning allows training machine learning models over decentralised data, it still relies on the centralised computation of model aggregations. In this presentation I will present recent research works targeting the decentralisation of machine learning beyond the well know Federated Learning concept. I will particularly focus on fundamental questions such as whether decentralisation increases or reduces the attack surface and whether decentralisation may improve personalisation compared to classical centralised cloud-based approaches.