BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VEVENT DTSTAMP:20260124T054049Z UID:EA43CEA8-7A6D-4730-9C04-1C763BABF117 DTSTART;TZID=Etc/UTC:20250116T120000 DTEND;TZID=Etc/UTC:20250116T130000 DESCRIPTION:Special Presentation by Dr. Mérouane Debbah (Khalifa U.\, UAE) \n\nHosted by the Future Networks Artificial Intelligence & Machine Learni ng (AIML) Working Group\n\nDate/Time: Thursday\, January 16th\, 2025 @ 12: 00 UTC\n\nTopic:\n\nGenerative Diffusion Models for Network Optimization\n \nAbstract:\n\nNetwork optimization is a fundamental challenge in Internet -of-Things (IoT) networks\, often characterized by complex features that m ake it difficult to solve these problems. Recently\, generative diffusion models (GDMs) have emerged as a promising new approach to network optimiza tion\, with the potential to directly address these optimization problems. However\, the application of GDMs in this field is still in its early sta ges\, and there is a noticeable lack of theoretical research and empirical findings. In this study\, we first explore the intrinsic characteristics of generative models. Next\, we provide a concise theoretical proof and in tuitive demonstration of the advantages of generative models over discrimi native models in network optimization. Based on this exploration\, we impl ement GDMs as optimizers aimed at learning high-quality solution distribut ions for given inputs\, sampling from these distributions during inference to approximate or achieve optimal solutions. Specifically\, we utilize de noising diffusion probabilistic models (DDPMs) and employ a classifier-fre e guidance mechanism to manage conditional guidance based on input paramet ers. We conduct extensive experiments across three challenging network opt imization problems. By investigating various model configurations and the principles of GDMs as optimizers\, we demonstrate the ability to overcome prediction errors and validate the convergence of generated solutions to o ptimal solutions.\n\nSpeaker:\n\nDr. Mérouane Debbah is a Professor at th e Khalifa University of Science and Technology in Abu Dhabi and founding D irector of the KU 6G Research Center. He is a frequent keynote speaker at international events in the field of telecommunication and AI. His researc h has been lying at the interface of fundamental mathematics\, algorithms\ , statistics\, information and communication sciences with a special focus on random matrix theory and learning algorithms. In the Communication fie ld\, he has been at the heart of the development of small cells (4G)\, Mas sive MIMO (5G) and Large Intelligent Surfaces (6G) technologies. In the AI field\, he is known for his work on Large Language Models\, distributed A I systems for networks and semantic communications. He received multiple p restigious distinctions\, prizes and best paper awards (more than 40 IEEE best paper awards) for his contributions to both fields and according to r esearch.com he is ranked as the best scientist in France in the field of E lectronics and Electrical Engineering. He is an IEEE Fellow\, a WWRF Fello w\, a Eurasip Fellow\, an AAIA Fellow\, an Institut Louis Bachelier Fellow \, an AIIA Fellow\, and a Membre émérite SEE. He is chair of the IEEE La rge Generative AI Models in Telecom (GenAINet) Emerging Technology Initiat ive and a member of the Marconi Prize Selection Advisory Committee.\n\nCo- sponsored by: Artificial Intelligence & Machine Learning (AIML) Working Gr oup\n\nVirtual: https://events.vtools.ieee.org/m/453702 LOCATION:Virtual: https://events.vtools.ieee.org/m/453702 ORGANIZER:c.polk@comsoc.org SEQUENCE:28 SUMMARY:Generative Diffusion Models for Network Optimization URL;VALUE=URI:https://events.vtools.ieee.org/m/453702 X-ALT-DESC:Description: <br /><p class="MsoNormal" style="margin-top: .25in \;"><img src="https://events.vtools.ieee.org/vtools_ui/media/display/5f4a6 ce8-dbd6-4684-9f04-311e88668911"></p>\n<p class="MsoNormal" style="margin- top: 12.0pt\;">Special Presentation by<strong> Dr. M&eacute\;rouane Debbah (Khalifa U.\, UAE)</strong></p>\n<p class="MsoNormal" style="margin-top: 12.0pt\;">Hosted by the Future Networks<strong> Artificial Intelligence &a mp\; Machine Learning (AIML) Working Group</strong></p>\n<p class="MsoNorm al" style="margin-top: 12.0pt\;"><strong><span style="font-size: 14.0pt\; font-family: Copperplate\; mso-fareast-font-family: PMingLiU\; mso-fareast -theme-font: minor-fareast\; mso-bidi-font-family: Arial\; mso-bidi-theme- font: minor-bidi\; mso-ansi-language: EN-US\; mso-fareast-language: ZH-TW\ ; mso-bidi-language: AR-SA\;">Date/Time</span></strong><span style="font-s ize: 12.0pt\; font-family: 'Calibri'\,sans-serif\; mso-ascii-theme-font: m inor-latin\; mso-fareast-font-family: PMingLiU\; mso-fareast-theme-font: m inor-fareast\; mso-hansi-theme-font: minor-latin\; mso-bidi-font-family: A rial\; mso-bidi-theme-font: minor-bidi\; mso-ansi-language: EN-US\; mso-fa reast-language: ZH-TW\; mso-bidi-language: AR-SA\;">: <strong>Thursday\, J anuary 16th\, 2025</strong><strong>&nbsp\;@ 12:00 UTC</strong></span></p>\ n<p class="MsoNormal" style="margin-top: .25in\;"><strong><u><span style=" font-size: 16.0pt\; font-family: Copperplate\;">Topic</span></u></strong>< strong><span style="font-size: 16.0pt\; font-family: Copperplate\;">:</spa n></strong></p>\n<p class="MsoNormal"><strong><span style="font-size: 16.0 pt\;">Generative Diffusion Models for Network Optimization</span></strong> </p>\n<p class="MsoNormal" style="margin-top: .25in\;"><strong><u><span st yle="font-size: 16.0pt\; font-family: Copperplate\;">Abstract</span></u></ strong><strong><span style="font-size: 16.0pt\; font-family: Copperplate\; ">:</span></strong></p>\n<p class="MsoNormal"><span style="font-size: 12.0 pt\; font-family: 'Calibri'\,sans-serif\; mso-ascii-theme-font: minor-lati n\; mso-fareast-font-family: PMingLiU\; mso-fareast-theme-font: minor-fare ast\; mso-hansi-theme-font: minor-latin\; mso-bidi-font-family: Arial\; ms o-bidi-theme-font: minor-bidi\; mso-ansi-language: EN-US\; mso-fareast-lan guage: ZH-TW\; mso-bidi-language: AR-SA\;">Network optimization is a funda mental challenge in Internet-of-Things (IoT) networks\, often characterize d by complex features that make it difficult to solve these problems. Rece ntly\, generative diffusion models (GDMs) have emerged as a promising new approach to network optimization\, with the potential to directly address these optimization problems. However\, the application of GDMs in this fie ld is still in its early stages\, and there is a noticeable lack of theore tical research and empirical findings. In this study\, we first explore th e intrinsic characteristics of generative models. Next\, we provide a conc ise theoretical proof and intuitive demonstration of the advantages of gen erative models over discriminative models in network optimization. Based o n this exploration\, we implement GDMs as optimizers aimed at learning hig h-quality solution distributions for given inputs\, sampling from these di stributions during inference to approximate or achieve optimal solutions. Specifically\, we utilize denoising diffusion probabilistic models (DDPMs) and employ a classifier-free guidance mechanism to manage conditional gui dance based on input parameters. We conduct extensive experiments across t hree challenging network optimization problems. By investigating various m odel configurations and the principles of GDMs as optimizers\, we demonstr ate the ability to overcome prediction errors and validate the convergence of generated solutions to optimal solutions.</span></p>\n<p><strong><span style="font-size: 16.0pt\; font-family: Copperplate\;"><u>Speaker</u>:</s pan></strong></p>\n<table style="border-collapse: collapse\; width: 100%\; " border="1"><colgroup><col style="width: 24.3015%\;"><col style="width: 7 5.7635%\;"></colgroup>\n<tbody>\n<tr>\n<td><img src="https://events.vtools .ieee.org/vtools_ui/media/display/d20008d6-0fba-4648-99c8-572006717a3b"></ td>\n<td>\n<p class="MsoNormal" style="margin-top: 6.0pt\;"><span style="m so-no-proof: yes\;">Dr. M&eacute\;rouane Debbah is a Professor at the Khal ifa University of Science and Technology in Abu Dhabi and founding Directo r of the KU 6G Research Center. He is a frequent keynote speaker at intern ational events in the field of telecommunication and AI. His research has been lying at the interface of fundamental mathematics\, algorithms\, stat istics\, information and communication sciences with a special focus on ra ndom matrix theory and learning algorithms. In the Communication field\, h e has been at the heart of the development of small cells (4G)\, Massive M IMO (5G) and Large Intelligent Surfaces (6G) technologies. In the AI field \, he is known for his work on Large Language Models\, distributed AI syst ems for networks and semantic communications. He received multiple prestig ious distinctions\, prizes and best paper awards (more than 40 IEEE best p aper awards) for his contributions to both fields and according to researc h.com he is ranked as the best scientist in France in the field of Electro nics and Electrical Engineering. He is an IEEE Fellow\, a WWRF Fellow\, a Eurasip Fellow\, an AAIA Fellow\, an Institut Louis Bachelier Fellow\, an AIIA Fellow\, and a Membre &eacute\;m&eacute\;rite SEE. He is chair of the IEEE Large Generative AI Models in Telecom (GenAINet) Emerging Technology Initiative and a member of the Marconi Prize Selection Advisory Committee .</span></p>\n<p class="MsoNormal" style="margin-top: 6.0pt\;">&nbsp\;</p> \n</td>\n</tr>\n</tbody>\n</table> END:VEVENT END:VCALENDAR