Edge intelligence, or edge-native artificial intelligence, is expected to become a key enabler for future 6G networks to fulfil the vision of ubiquitous intelligence. Data-driven artificial intelligence combined with communications networks and mobile edge computing will bring new functionalities and services that were missing in 5G networks, enabling an enhanced quality of experience for users. Countless applications will benefit from edge intelligence, including novel personal smart device environments and experiences with augmented reality, holographic projection, remote surgery, smart manufacturing, urban computing, and autonomous driving, to name a few.
Open Day CTM 2021 aims to showcase a short selection of the Centre for Telecommunications and Multimedia’s research activities in the different domains of edge intelligence, while also bringing together a range of Key Stakeholders from industry and academia to present their views and discuss promising approaches addressing the implementation challenges of edge intelligence for 6G networks.
Watch on YouTubeAbstract: Neuromorphic (brain-like) photonics is attracting increasing interest for new paradigms in computing and artificial intelligence. In this talk, I review our research on light-enabled neuromorphic photonic systems built with artificial laser neurons, and describe their application for information processing functionalities and neuronal circuit emulation at ultrafast speeds.
About Antonio Hurtado: Dr Antonio Hurtado is a Turing AI Fellow and Senior Lecturer (Associate Professor) at the University of Strathclyde’s Institute of Photonics (IoP) where he currently leads research programmes on neuromorphic photonics. He is the Principal Investigator (PI) for the ‘BrainLaser’ project funded by the USA’s Office of Naval Research Global (ONR Global) as well as Strathclyde’s PI for the EU FET Open Project ‘ChipAI’. In 2020 he was awarded a Turing AI Acceleration Fellowship by the UK Governement to develop a 5-year research programme in “Photonics for Ultrafast Artificial Intelligence (PHOTON-AI)”.
Abstract: Monitoring of physiological signals, in particular those that provide information about the heart, is attracting interest for a myriad of innovative applications. Nowadays, it is possible to ubiquitously monitor a user by means of several technologies, like PPG, ECG, and others. In light of CardioID's latest developments, this talk will review how it is possible to integrate these technologies in the industry, showcasing application scenarios for continuous biometric recognition and driver monitoring.
About André Lourenço: André Lourenço is the CEO and Co-founder of CardioID Technologies. At CardioID Technologies, he is converting ECG biometrics research into an innovative business. He is applying the knowledge and experience gathered while founding Minalytics (software analytics and data mining startup, 2013), while working for startups Albatroz Engineering (software and hardware development, 2006) and Lusospace (space technology development, 2003-2005), and conducting research on Pattern and Image Analysis over +10 years at Instituto de Telecomunicações (IT) – Instituto Superior Técnico (IST), including numerous projects funded by the National Science Foundation (FCT). André holds Licenciatura, M.Sc. and PhD degrees, all three in Electrical and Computer Engineering from the Instituto Superior Técnico (IST), Universidade de Lisboa (UL), Portugal. Invited Professor at ISEL since 2005, André also lectures information processing and programming and has supervised several research projects, 6 M.Sc. theses, and 27 final degree projects. Scientific contributions include 12 book chapters, 8 peer-reviewed papers, 77 international conference papers and 1 patent-pending – “Device and Method for Continuous Biometric Recognition Based on Electrocardiographic Signals”.
Abstract: Edge AI is now established as one of the key enablers of 6G, and has received significant attention in both academic and industry. This short talk will first overview the current state of the art techniques in distributed and communication-efficient ML over wireless (such as federated learning/distillation and split learning). Then the talk will provide a glimpse of what the next frontier in edge AI is.
About Mehdi Bennis: Dr Mehdi Bennis is an Associate Professor at the Centre for Wireless Communications, University of Oulu, Finland, Academy of Finland Research Fellow and head of the intelligent connectivity and networks/systems group (ICON). His main research interests are in radio resource management, heterogeneous networks, game theory and distributed machine learning in 5G networks and beyond. He has published more than 200 research papers in international conferences, journals and book chapters. He has been the recipient of several prestigious awards including the 2015 Fred W. Ellersick Prize from the IEEE Communications Society, the 2016 Best Tutorial Prize from the IEEE Communications Society, the 2017 EURASIP Best paper Award for the Journal of Wireless Communications and Networks, the all-University of Oulu award for research, the 2019 IEEE ComSoc Radio Communications Committee Early Achievement Award and the 2020 Clarviate Highly Cited Researcher by the Web of Science. Dr Bennis is an editor of IEEE TCOM and Specialty Chief Editor for Data Science for Communications in the Frontiers in Communications and Networks journal. Dr Bennis is an IEEE Fellow.
Abstract: This talk will outline the merits of edge computing for emerging IoT and Artificial Intelligence Applications. Accordingly, it will present different paradigms for deploying and executing AI at the edge (e.g., Federated Machine Learning, Embedded Machine Learning, TinyML) along with the role of the networking infrastructure (4G/5G/6G, MEC). The pros and cons of each different approach will be given and application examples will be illustrated.
About John Soldatos: John Soldatos holds a PhD in Electrical & Computer Engineering from the National Technical University of Athens (2000) and is currently Honorary Research Fellow at the University of Glasgow, UK. He was Associate Professor and Head of the Internet of Things (IoT) Group at the Athens Information Technology (AIT), Greece (2006–2019), and Adjunct Professor at the Carnegie Mellon University, Pittsburgh, PA (2007–2010). Since January 2020 he is a Senior R&D Consultant Innovation Delivery Specialist with INTRASOFT International. He has significant experience in working closely with large multi-national industries, while being scientific advisor to high-tech startup enterprises. Dr. Soldatos is an expert in Internet-of-Things (IoT) and Artificial Intelligence, with experience in industry (Industry 4.0), finance and healthcare applications. He has played a leading role in the delivery of more than 60 (industrial, research, and consulting) projects, for both private & public sector organizations. He is co-founder of the OpenIoT project and has published over 200 articles in international journals, books, and conference proceedings. He has received national and international recognition through appointments in standardization working groups, expert groups, and various boards. Dr. Soldatos has coedited and co-authored three edited volumes (books) on Internet of Things topics, including IoT for Industrial Automation, IoT Analytics, and IoT Security. He is the author of the book “A 360 Degrees View of IoT Technologies” (Artech House, December 2020).
About Manuel Ricardo: Manuel Ricardo holds a PhD (2000) degree in Electrical and Computers Engineering (ECE, (Telecommunications) from University of Porto (FEUP). He his professor Catedrático (full professor) at FEUP where he teaches courses on Computer Networks and Wireless Networks. At FEUP he is also member of the Executive Committee of his department (ECE) and member of the Scientific Committee of the Doctoral Program in ECE. At INESC TEC he is member of the board of directors and leads the research cluster on Networked Intelligent Systems. His research interests include radio resource management, network congestion control, wireless networks, quality of service, and performance evaluation. Manuel Ricardo has published 120+ research papers and participated in 30+ research projects.
Abstract: In this talk, we will introduce the IoMMT concept and compare it to IoT. We will make reference to typical IoMMT objects and to the requirements they impose in terms of bandwidth, computing resources and security. Real world use cases associated to IoMMT will be presented. On-going standardisation efforts, existing solutions and still open challenges will be addressed.
About Teresa Andrade: Maria Teresa Andrade is an Assistant Professor at FEUP, in the DEEC. She obtained a degree in Electrotechnical and Computing Engineering in 1986, the MSc in 1992 and the PhD in 2008, at FEUP. She participates in research activities at INESC TEC, integrated in the Multimedia Systems Area. Main interests include context-awareness; mobile and adaptable multimedia applications in heterogeneous environments; 3D and multiview video streaming; quality of service and of experience in multimedia services; semantic technologies and content recommendation; digital television, digital cinema and new media; Internet of Multimedia Things (IoMMT).
Abstract: AI algorithms, big data, communication networks and powerful computing hardware have all collectively enabled the modern age of AI. But all breakthroughs cast new challenges, and leveraging the expenditure of energy and data, of present day systems, with new hardware and algorithm paradigms is fundamental to bring AI to the next stage. The idea that in-memory computing could be a key to dramatically improve efficiency in both grounds is becoming widely accepted. This presentation will then present the developments of the project NeurOxide that aims to bring low-cost in-memory computing closer to reality, by integrating memory nano-devices (memristor type) together with transistors on a single technology.
About Vítor Tavares: Vítor Grade Tavares earned his undergraduate and MSc degrees from the University of Aveiro, Portugal, and the Ph.D. degree from the Computational NeuroEngineering Laboratory, University of Florida, Gainesville, USA, in 2001, all in electrical engineering. He is currently an Assistant Professor at University of Porto and a Senior Researcher at INESC-TEC, Porto. In 2010 he was a Visiting Professor at Carnegie Mellon University, USA. His research interests include low-power, mixed-signal and neuromorphic integrated-chip design and biomimetic computing, CMOS RF integrated circuit design for wireless sensor networks, and large electronics. He has coordinated several national projects, and also locally coordinated European projects. He was awarded with a certificate of appreciation for contributions towards the advancement of IEEE and the Engineering Professions as Chair of the Education Society Chapter - Portugal section, which he co-founded, and presently is the Vice-Chair.
Abstract: The shift of data to the Cloud brought many advantages, however, as the IoT universe expands, Cloud computing will not be up to the task of processing the mountains of data being generated by the increasing number of devices and other digital platforms connected to the various enterprise networks. Now that almost everything that can be centralized has been centralized, most of the new opportunities for the “cloud” lie at the “edge”. Edge computing is computing that is done at or near the source of the data, instead of relying on the Cloud. In a simple way, edge computing brings the data and the compute closest to the point of interaction. In this talk, we will go through several examples of edge computing for biometrics and wellbeing. For example, nowadays some of the most advanced biometric solutions can enrol, capture, compare, match and process data all on their own. This data can be used in a number of applications dealing with extremely accurate data and reporting in real-time. Beyond traditional access control, it has high potential for the detection of counterfeit biometric data or the classification of psychophysiological states as drowsiness, stress. Environments such as border control applications also gain from the ease and quickness of processing times provided by edge computing as well as real time analysis of compliance to the requirements of video or image data.
About Ana F. Sequeira: Ana F. Sequeira holds a Licenciatura (5-year degree) in Mathematics, a MSc in Mathematical Engineering and a PhD in Electrical and Computer Engineering, all from the University of Porto. Sequeira is an assistant research at INESC TEC, where she is associated with the Visual Computing & Machine Intelligence (VCMI) research group. Sequeira collaborated on EU FP7 and H2020 projects (FASTPASS and PROTECT); worked with IrisGuard UK developing an anti-spoofing measure for the EyePay Technology. Sequeira led the construction of several biometric databases; managed biometric competitions and co-authored several research publications. Her main research interests include Machine Learning and Computer Vision for Biometrics; Presentation Attack Detection, Mobile Biometrics, Border Control and XAI.
Abstract: Aerial networks, composed of Unmanned Aerial Vehicles (UAVs) acting as Wi-Fi access points or cellular base stations, are emerging as a solution to provide on-demand wireless connectivity to users in extreme scenarios. This presentation explores the use of deep reinforcement learning to design a traffic-aware UAV placement algorithm that improves the performance of aerial wireless networks.
About Eduardo N. Almeida: Eduardo N. Almeida is an R&D Engineer at INESC TEC and a PhD candidate in the Doctoral Programme in Telecommunications (MAP-tele) at Faculdade de Engenharia, Universidade do Porto (FEUP), Portugal. Currently, he is pursuing his PhD research on traffic-aware Unmanned Aerial Vehicle (UAV) placement for aerial wireless networks. He has participated in national and international R&D projects at INESC TEC, including WISE and ResponDrone. His main research interests include aerial wireless networks, UAV placement and reinforcement learning.
| May 4th, 2021 All times on Western European Summer Time (UTC+1, Lisbon time) |
|||||||||
|---|---|---|---|---|---|---|---|---|---|
| 14:00 | Opening Session CTM Coordination |
||||||||
| 14:20 | CTM Vision
|
||||||||
| 15:00 | Break | ||||||||
| 15:15 | Key Stakeholders' Vision
|
||||||||
| 16:15 | Panel Discussion with Key Stakeholders Moderated by Manuel Ricardo (INESC TEC and FEUP) |
||||||||
| 17:00 | Closing Session Manuel Ricardo (INESC TEC and FEUP) |
Registration is now closed.
Thanks to all attendees for joining us.
Stay tuned for Open Day CTM 2022!
