Events

EXPLAINABLE AND ROBUST AI FOR 6G Dr. SINEM COLERI When: November 21st at 10H30 AM EST Via zoom: https://uqtr.zoom.us/j/81521084215?pwd=bchQDndZg7DTlpVuaeag6bhGwaOvn9.1 Meeting ID: 815 2108 4215 Password: 018477 Unlike previous generations of wireless networks, which were primarily designed to meet the requirements of human communications, 5G networks enable extensive data collection from machines. As we transition to 6G, the emphasis moves beyond connectivity toward leveraging this machine-generated data for a new spectrum of control applications, such as UAV swarms, collaborative robots, and cooperative autonomous vehicles. Designing communication systems for these advanced control applications introduces a distinct set of challenges. These include meeting stringent requirements for delay and reliability, addressing the semantics of control systems, and ensuring robust resource management. In the first part of this talk, we explore ultra-reliable channel modeling and communication techniques based on the integration of extreme value theory with generative artificial intelligence (AI). These methods offer improved accuracy in predicting rare but critical events while providing adaptivity to dynamic scenarios. In the second part of the talk, we explore the benefits of employing optimization theory based, explainable, and robust AI in radio resource management for the joint design of control and communication systems. These approaches offer a systematic methodology to enhance robustness and interpret decisions made by black-box AI models. BIOGRAPHY Dr. Sinem Coleri is a Professor in the Department of Electrical and Electronics Engineering at Koc University. She is also the founding director of Wireless Networks Laboratory (WNL) and director of Ford Otosan Automotive Technologies Laboratory. Sinem Coleri received the BS degree in electrical and electronics engineering from Bilkent University in 2000, the M.S. and Ph.D. degrees in electrical engineering and computer sciences from University of California Berkeley in 2002 and 2005. She worked as a research scientist in Wireless Sensor Networks Berkeley Lab under sponsorship of Pirelli and Telecom Italia from 2006 to 2009. Since September 2009, she has been a faculty member in the department of Electrical and Electronics Engineering at Koc University. Her research interests are in 6G wireless communications and networking, AI-based wireless networks, machine-to-machine communications, wireless networked control systems and vehicular networks. Dr. Coleri has more than 150 publications with citations over 12000 (Google scholar profile). She has received numerous awards and recognitions, including TUBITAK (The Scientific and Technological Research Council of Turkey) Science Award in 2024; N2Women: Stars in Computer Networking and Communications in 2022; TUBITAK Incentive Award and IEEE Vehicular Technology Society Neal Shepherd Memorial Best Propagation Paper Award in 2020; Outstanding Achievement Award by Higher Education Council in 2018; and Turkish Academy of Sciences Distinguished Young Scientist (TUBA-GEBIP) Award in 2015. Dr. Coleri currently holds the position of Editor-in-Chief at the IEEE Open Journal of the Communications Society. Dr. Coleri is an IEEE Fellow, AAIA Fellow and IEEE ComSoc Distinguished Lecturer. Speaker(s): Dr. Coleri Virtual: https://events.vtools.ieee.org/m/509632

Despite the rapid advances in Large Language Models (LLM), research efforts have historically focused disproportionately on high-resource languages, particularly English, leaving over 7,000 living languages underserved. We address the fundamental challenge of bridging the gap of low-resource language (LRL) translation in multilingual language models. Low-resource languages are typically characterized by a scarcity of both unlabeled and labeled data, as well as limited tools and models. This talk explores strategies aimed at bridging the gap of low-resource language (LRL) translation in multilingual models, where LRLs are characterised by a limited scarcity of both unlabeled and labelled data, as well as limited tools and models. Speaker(s): Dr. Lee, Room: 313, Bldg: Bergeron Centre for Engineering Excellence, 11 Arboretum Ln, North York, Ontario, Canada, M3N 3A7

ABSTRACT: The compute demands of AI and robotics continue to rise due to the rapidly growing volume of data to be processed; the increasingly complex algorithms for higher quality of results; and the demands for energy efficiency and real-time performance. In this talk, we will discuss the design of efficient tailored hardware accelerators and the co-design of algorithms and hardware that reduce the energy consumption while delivering swift real-time and robust performance for applications including deep neural networks, data analytics with sparse tensor algebra, and autonomous navigation. Throughout the talk, we will highlight important design principles, methodologies, and tools that can facilitate an effective design process and various forms of co-design that can broaden the design space. BIO: Vivienne Sze is a professor in the Electrical Engineering and Computer Science Department at MIT. She works on computing systems that enable energy-efficient machine learning, computer vision, and video compression/processing for a wide range of applications, including autonomous navigation, digital health, and the internet of things. She is widely recognized for her leading work in these areas and has received awards, including faculty awards from Google, Facebook, and Qualcomm, the Symposium on VLSI Circuits Best Student Paper Award, the IEEE Custom Integrated Circuits Conference Outstanding Invited Paper Award, and the IEEE Micro Top Picks Award. As a member of the Joint Collaborative Team on Video Coding, she received the Primetime Engineering Emmy Award for the development of the High-Efficiency Video Coding video compression standard. She is a co-editor of High Efficiency Video Coding (HEVC): Algorithms and Architectures (Springer, 2014) and co-author of Efficient Processing of Deep Neural Networks (Synthesis Lectures on Computer Architecture, Morgan Claypool, 2020). For more information about Prof. Sze’s research, please visit (https://can01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fsze.mit.edu%2F&data=05%7C02%7Ckelly.hunter%40mail.utoronto.ca%7C0c1d2bb1b79a45865a2f08de1bb0ba44%7C78aac2262f034b4d9037b46d56c55210%7C0%7C0%7C638978643061187412%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=rBYJubhz5nDvA8lPUSMBcl2TGsc9ZebyegOiKWWoVtg%3D&reserved=0). Speaker(s): Vivienne Room: MC252, Bldg: Mechanical Engineering Building, 5 King's College Road, Toronto, Ontario, Canada, M5S3G8

Software-Defined Optics: Opening New Frontiers in Perception and Autonomy From autonomous vehicles to smart infrastructure, the future of intelligent systems depends on sensors that can adapt, learn, and perceive with precision. Lumotive is pioneering this transformation with light-control metasurface (LCM) technology, a revolutionary solid-state beam-steering approach that replaces mechanical LiDAR with fully programmable, software-defined optics. In this talk, Dr. Calvin Cheng, Director of Applications and Customer Success at Lumotive, will explore how LCM-based LiDAR systems enable dynamic scanning, on-device intelligence, and efficient 3D perception across diverse applications. Through live demonstrations and system-level insights, he will illustrate how Lumotive’s innovations are redefining the boundaries of perception and autonomy and how collaboration between academia and industry can accelerate the next generation of intelligent sensing systems. Speaker(s): Calvin Cheng PhD Room: 3038, Bldg: Macleod building, 2356 Main Mall, Vancouver, British Columbia, Canada, V6T 1Z4, Virtual: https://events.vtools.ieee.org/m/514914

Photonic Pathways to Scalable and Low-Cost Free-Space Optical Satellite Downlinks Abstract: Free-space optical communication (FSOC) is rapidly emerging as a cornerstone of high-bandwidth, low-latency connectivity for space-to-ground and inter-satellite networks. Yet the downlink to ground remains the most technically demanding segment, where atmospheric turbulence distorts optical wavefronts and drives system complexity and cost. Traditional bulk-optics solutions are difficult to align, power-hungry, and expensive to reproduce. Photonic technologies offer a transformative alternative: compact, robust, and inherently manufacturable architectures that merge adaptive optics, beam combination, and phase control directly on a chip. These integrated systems enable ultra-high-speed, real-time correction of turbulence with minimal mass and power, while leveraging semiconductor fabrication to achieve wafer-scale replication and dramatically lower cost. This cost reduction opens the possibility of deploying many more optical ground stations, enhancing global network coverage, redundancy, and resilience. By uniting astronomical adaptive optics with scalable photonic integration, we are charting a path toward affordable, high-performance optical downlinks to ground. I will highlight recent prototype demonstrations and outline our roadmap toward fully photonic optical ground terminals that bring the precision of astronomical instrumentation to the future of space communications. ------------------------------------------------------------------------ Voies photoniques pour des liaisons descendantes optiques par satellite en espace libre, évolutives et peu coûteuses Résumé: La communication optique en espace libre (FSOC) s'impose rapidement comme un pilier de la connectivité à haut débit et faible latence pour les réseaux sol-espace et inter-satellites. Cependant, la liaison descendante vers le sol demeure le segment le plus complexe sur le plan technique, car la turbulence atmosphérique y déforme les fronts d'onde optiques, ce qui accroît la complexité et le coût du système. Les solutions optiques classiques sont difficiles à aligner, énergivores et coûteuses à reproduire. Les technologies photoniques offrent une alternative révolutionnaire : des architectures compactes, robustes et intrinsèquement industrialisables qui intègrent l'optique adaptative, la combinaison de faisceaux et le contrôle de phase directement sur une puce. Ces systèmes intégrés permettent une correction ultrarapide et en temps réel de la turbulence avec une masse et une consommation d'énergie minimales, tout en tirant parti de la fabrication des semi-conducteurs pour une réplication à l'échelle de la plaquette et une réduction drastique des coûts. Cette réduction des coûts ouvre la voie au déploiement d'un nombre beaucoup plus important de stations optiques au sol, améliorant ainsi la couverture, la redondance et la résilience du réseau mondial. En associant l'optique adaptative astronomique à l'intégration photonique à grande échelle, nous ouvrons la voie à des liaisons optiques terrestres performantes et abordables. Je présenterai des démonstrations récentes de prototypes et décrirai notre feuille de route vers des terminaux optiques terrestres entièrement photoniques, qui mettront la précision des instruments astronomiques au service des communications spatiales de demain. Prof. Suresh Sivanandam, University of Toronto About / A propos The High Throughput and Secure Networks (HTSN) Challenge program is hosting regular virtual seminar series to promote scientific information sharing, discussions, and interactions between researchers. https://nrc.canada.ca/en/research-development/research-collaboration/programs/high-throughput-secure-networks-challenge-program Le programme Réseaux Sécurisés à Haut Débit (RSHD) organise régulièrement des séries de séminaires virtuels pour promouvoir le partage d’informations scientifiques, les discussions et les interactions entre chercheurs. https://nrc.canada.ca/fr/recherche-developpement/recherche-collaboration/programmes/programme-defi-reseaux-securises-haut-debit NEW: In order to promote more open discussions/interactions, at the end of the presentation and Q/A, we will allow other experts in this field (quantum comm) to present very briefly their work (1 slide, 2 min max) or their company. / Afin de favoriser des discussions/interactions plus ouvertes, à la fin de la présentation et des questions/réponses, nous permettrons aux experts de ce domaine (communications quantiques) de présenter très brièvement leurs travaux (1 diapositive, 2 min max) ou leur compagnie. Co-sponsored by: National Research Council, Canada. Speaker(s): Prof. Suresh Sivanandam, Virtual: https://events.vtools.ieee.org/m/514232