Keynotes

Collaborative In-Network Intelligence: Mechanisms for a Data Plane that Sees, Thinks, and Cooperates
Posco (Fung Po) Tso is a Professor of Computer Networks and Systems at Loughborough University, where his research focuses on making computer networks smarter — enabling them not just to carry data, but to process and act on it in real time. A central theme of his work is bringing AI and computing to the very edge of the network, close to where data is generated. His research has made the leap from lab to market: his ServerlessEdge runtime has been integrated into a commercial 5G edge router, the result of three successive jointly-funded projects spanning over five years. His 5G and Internet-of-Things research also underpins a live sensing deployment in the town of Bordon that continuously streams environmental data for ongoing research. He leads the IoDT^2 project, a €2.68M European consortium exploring the Internet of Digital Twin Things, and has secured over £5M in competitive research funding from UK and European sources including EPSRC, Innovate UK, and the UK Government’s SBRI programme. His international research collaborations span institutions in the UK, China, and beyond, with results published at leading computing venues. 
 
Posco is a Senior Member of the IEEE, an Area Editor of the journal Computer Networks, and serves on the UKRI Talent Peer Review College, which advises on national research fellowship awards. He is also a Google Cloud Faculty Expert and a GitHub Campus Advisor. He has published over 80 papers at leading IEEE and ACM venues including INFOCOM, ICNP, IEEE/ACM Transactions on Networking, IEEE Transactions on Services Computing, and IEEE Transactions on Parallel and Distributed Systems.
The network has always been the most intimate observer of its own traffic — yet for decades, intelligence has lived elsewhere: in servers, controllers, and management planes reached only after costly round-trips. What if the forwarding hardware itself could sense the network with surgical precision, reason about what it finds, and coordinate a response, all without leaving the data plane?
 
This keynote is organised around three interlocking capabilities. Seeing — collecting telemetry that is accurate, bandwidth-efficient, and resilient to loss, by redesigning the encoding and placement of measurement logic as first-class algorithmic problems. Thinking — running machine-learning inference at line rate, directly inside the switch pipeline, by co-designing neural architectures and compression schemes with the harsh constraints of the silicon: no floating point, no native multiply, a handful of pipeline stages. Cooperating — deploying these intelligent functions across multiple switches and planes as a joint optimisation, so that the network as a whole achieves what no single device can.
 
Each capability is technically demanding on its own and together they form a closed loop. Better telemetry feeds better inference; better inference sharpens the decisions that determine what is worth measuring next; and principled cross-switch coordination ensures that the resulting workload is placed where resources allow. The unifying insight across all three thrusts is that the algorithm must be co-designed with the hardware — and that doing so consistently unlocks capabilities assumed to be out of reach on forwarding silicon.
Posco Fung Po Tso

Professor of Networks and Systems, Department of Computer Science, Loughborough University, UK

The Road Less Hacked: Enhancing Security in Automotive Systems

Xiapu Luo is a professor in the Department of Computing, Associate Dean (Research) of the Faculty of Computer and Mathematical Sciences, and Co-Director of the Research Centre for Blockchain Technology at the Hong Kong Polytechnic University. His research focuses on Blockchain and Smart Contracts Security, Mobile and IoT Security, Network Security and Privacy, and Software Engineering with papers published in top venues. His research led to more than ten best/distinguished paper awards, including ACM CCS’24 Distinguished Paper Award, four ACM SIGSOFT Distinguished Paper Awards, Best DeFi Papers Award 2023, Best Paper Award in INFOCOM’18, Best Research Paper Award in ISSRE’16, etc. and several awards from the industry. He received the BOCHK Science and Technology Innovation Prize (FinTech) 2023 for his contribution to blockchain security. He also received Gold Award at the Cyber Security Professional Awards (Government Departments & Public Bodies Sector) 2025. He actively contributes to the community by participating in program/organization committees for major conferences in security, software engineering, and networking, and by serving as an associate editor for prestigious journals such as IEEE/ACM Transactions on Networking (TNET), IEEE Transactions on Dependable and Secure Computing (TDSC), IEEE Transactions on Network and Service Management (TNSM), and ACM Transactions on Privacy and Security (TOPS). He is also a member of the Hong Kong Monetary Authority’s CBDC Expert Group.

As vehicles transform into complex, interconnected systems, cybersecurity emerges as a paramount concern. This presentation will begin with an examination of the current automotive attack surfaces. We will then discuss our research on identifying vulnerabilities in critical automotive components and provide an overview of the defense mechanisms we have designed.

Prof. Xiapu Luo

Assoc. Dean & Professor, Hong Kong Polytechnic University

This event is a part of EAI Conference series
Join Free
Become an EAI Member
© EAI CollaborateCom 2026
Cookies Policy
Terms, Conditions and Privacy Policy
Scroll to Top