Sixth UK Mobile, Wearable and Ubiquitous Systems Research Symposium

Invited Speakers

Noa Zilberman (University of Oxford)

Title: Edge-Powered Dynamic and Secure Swarm Networking

Abstract: Smart environments managing dynamic swarms of mobile nodes often face strong requirements on latency, resilience, security and scalability. In this talk, I will present our work on SmartEdge, enabling decentralized edge intelligence for smart IoT applications in near real-time, and utilizing hardware-accelerated in-network operations for context-aware swarm networking. The talk will focus on the provision of ultra-low latency, ML driven in-network defence against emerging threats in dynamic swarm environments. It will introduce Planter, an open source framework for in-network ML, and the provision of distributed, federated and hybrid deployment of attack detection and mitigation within resource constrained network devices. Using examples from smart transportation and smart manufacturing, the talk will discuss some of the challenges and solutions to future development of smart mobile environments.

Bio: Noa Zilberman is an Associate Professor at the University of Oxford, where she leads the Computing Infrastructure group. Her research focuses on the integration of micro-level architectures and macro level, large scale systems.

Aiden Doherty (University of Oxford)

Title: Using wearables in large-scale biobanks to transform our understanding of health

Abstract: I will discuss the story behind the collection of wrist-worn accelerometer data in over 150,000 research participants across the UK and China, while also describing efforts to collect complementary open human activity recognition validation datasets to further enhance these resources. I will share the development of machine learning methods for sleep, sedentary behaviour, physical activity behaviours and steps, referring to open software tools and data resources of relevance to others in the field.

Bio: Aiden is a Wellcome Trust Senior Research Fellow and Professor of Biomedical Informatics at the University of Oxford.

Stylianos I. Venieris (Samsung AI Center, Cambridge, UK)

Title: A Call to Rethink AI Computing at the Consumer Edge: New Challenges and Systems Considerations

Abstract: In the last few years, the rapid progress of deep learning and deep neural networks (DNNs) has enabled the embedding of intelligence across consumer devices, be it voice assistants, smart cameras, or home robots. Nonetheless, recent trends strongly indicate that the next decade of consumer intelligence will require unprecedented levels of computational resources in order to cope with the demands of the new AI use-cases. In this talk, we argue for a paradigm shift towards the next generation of Consumer Edge-AI Computing. We'll start by discussing the new computational challenges of next-generation AI systems. Next, we'll introduce the notion of among-device intelligence, where multiple devices collaborate with each other through the fluid sharing of both context information and computational resources. Finally, we'll discuss how novel components, such as adaptive neural models, multi-DNN accelerators and fluid batching schemes, can be the key towards bringing performant and efficient intelligence to the consumer edge.

Bio: Stylianos Venieris is a Senior Research Scientist at Samsung AI and leads the Distributed AI group, focusing on on-device and distributed forms of machine learning.

Steve Beeby (University of Southampton)

Title: E-Textiles – a New Platform for Wearable Technology

Abstract: This talk will introduce electronic textiles (e-textiles or smart fabrics) and describe the progression of the technology towards becoming a viable platform for wearables. The ultimate vision of the technology is to enable user to engage with their wearable technologies by simply getting dress with the electronic functionality being imperceptible to the wearer. However, combining electronics with textiles is not straightforward due to the particular mechanical characteristics that enable fabrics to drape and flex as well as being very strong and soft and able to survive very harsh conditions. The talk will discuss the fabrication of E-textiles through the Steve Beeby holds a prestigious Royal Academy of Engineering Chair in Emerging Technologies on e-textile engineering. He completed his PhD studies in MEMS Resonant Sensors at the University of Southampton, in 1996 and he is a Professor at the University of Southampton since 2011. His research interests include the application of flexible electronics, smart printable materials, and energy-harvesting technologies to electronic textiles (e-textiles). He leads the E-Textiles Network and has established the E-Textiles International Conference series. He is a Fellow of the IEEE, IET and IoP. use of printed smart materials or the integration of flexible electronics and discuss the challenges of suppling power, scaling up towards mass manufacture and surviving the rigours of use.

Bio: Steve Beeby holds a prestigious Royal Academy of Engineering Chair in Emerging Technologies on e-textile engineering. He completed his PhD studies in MEMS Resonant Sensors at the University of Southampton, in 1996 and he is a Professor at the University of Southampton since 2011. His research interests include the application of flexible electronics, smart printable materials, and energy-harvesting technologies to electronic textiles (e-textiles). He leads the E-Textiles Network and has established the E-Textiles International Conference series. He is a Fellow of the IEEE, IET and IoP.

Javier Fernandez-Marques (Flower Labs)

Title: Federating everything with Flower

Abstract: Federated Learning has quickly become the preferred form of training AI models when the data cannot be moved to a centralised location due to privacy reasons, legal reasons, logistical reasons, or combinations of these. This talk will be divided into two parts: first, an overview of how a typical Federated Learning pipeline works, followed up by several real-world scenarios where FL has proven to be indispensable. We’ll close this first part of the talk by covering the current open research questions in FL with a special focus on those that arise in cross-device setups. The second part of the talk will be in the form of a tutorial where you will learn how to get started with FL using Flower. You will learn how to federate your existing ML projects and how to make the most out of Flower.

Bio: Javier is a research scientist at Flower Labs. He works on the core framework and develops the Flower Simulation Engine. Javier interests lie in the intersection of Machine Learning and Systems, and more concretely running on-device ML workloads efficiently, a key component in Federated Learning. Javier got his PhD in Computer Science from the University of Oxford in 2021. Before joining Flower Labs, he was a research scientist at Samsung AI (Cambridge, UK).

Organising Committee

• Alastair Beresford (University of Cambridge)
• Jagmohan Chahuan (University of Southampton)
• Katayoun Farrahi (University of Southampton)
• Shelly Vishwakarma (University of Southampton)
• Chong Tang (University of Southampton)
• Neelam Singh (University of Southampton)

Steering Committee

• Alastair Beresford (University of Cambridge)
• Sarah Clinch (University of Manchester)
• Nigel Davies (University of Lancaster)
• Christos Efstratiou (University of Kent)
• Katayoun Farrahi (University of Southampton)
• Fahim Kawsar (Nokia Bell Labs / University of Glasgow)
• Nic Lane (University of Cambridge)
• Mahesh Marina (University of Edinburgh)
• Cecilia Mascolo (University of Cambridge)
• Mirco Musolesi (University College London)
• Paul Patras (University of Edinburgh)
• Alex Rogers (University of Oxford)
• George Roussos (Birkbeck College)

Sponsors