BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Multi-Agent Graph Learning-Based Optimization and its Applications
  to Computer Networks - Guillermo Bernárdez Gil\, UPC Barcelona
DTSTART:20231108T170000Z
DTEND:20231108T180000Z
UID:TALK208117@talks.cam.ac.uk
CONTACT:Pietro Lio
DESCRIPTION:In the age of digital transformation\, network applications ar
 e placing unprecedented demands on network infrastructure. Traditional net
 work management solutions are simply not up to the task\, struggling to ke
 ep pace with the evolving needs of modern applications such as augmented a
 nd virtual reality\, holographic telepresence\, etc.\n\nMAGNNETO\, an inno
 vative Machine Learning framework for autonomous network optimization\, is
  designed to address the main operational challenges of modern networked s
 ystems. Our proposed framework integrates a Graph Neural Network architect
 ure into a Multi-Agent Reinforcement Learning setting\, enabling fully dec
 entralized optimization that capitalizes on the inherent distributed natur
 e of networked environments.\n\nMAGNNETO is adaptable and versatile\, offe
 ring the potential to handle a wide range of network optimization challeng
 es. It has already been successfully implemented in two highly impactful s
 cenarios: Traffic Engineering optimization and Congestion Control in Datac
 enters. In both cases\, MAGNNETO has achieved similar performance to state
 -of-the-art approaches\, offering significant improvements in execution co
 st while being totally compliant with current and legacy equipment.\n\nIn 
 this talk\, we will explore the MAGNNETO framework in detail\, discussing 
 its innovative design\, key features\, and multifaceted applications. In a
 ddition to this\, we will also introduce a promising research line to enha
 nce the framework by integrating Topological Deep Learning techniques\, en
 visioning a novel approach that goes beyond the traditional graph domain.\
 n\nJoin us as we embark on a pioneering journey to redefine the landscape 
 of distributed network optimization and management with MAGNNETO.\n\nLink 
 to the recorded talk: https://drive.google.com/file/d/1bDpFDyEoLNOwRg4CU5U
 K6yVaMqmwq1rw/view?usp=sharing
LOCATION:Lecture Theatre 2\, Computer Laboratory\, William Gates Building
END:VEVENT
END:VCALENDAR