Multi-Agent Graph Learning-Based Optimization and its Applications to Computer Networks

Add to your list(s) Download to your calendar using vCal

• Guillermo Bernárdez Gil, UPC Barcelona
• Wednesday 08 November 2023, 17:00-18:00
• Lecture Theatre 2, Computer Laboratory, William Gates Building.

If you have a question about this talk, please contact Pietro Lio.

In the age of digital transformation, network applications are placing unprecedented demands on network infrastructure. Traditional network management solutions are simply not up to the task, struggling to keep pace with the evolving needs of modern applications such as augmented and virtual reality, holographic telepresence, etc.

MAGNNETO , an innovative Machine Learning framework for autonomous network optimization, is designed to address the main operational challenges of modern networked systems. Our proposed framework integrates a Graph Neural Network architecture into a Multi-Agent Reinforcement Learning setting, enabling fully decentralized optimization that capitalizes on the inherent distributed nature of networked environments.

MAGNNETO is adaptable and versatile, offering the potential to handle a wide range of network optimization challenges. It has already been successfully implemented in two highly impactful scenarios: Traffic Engineering optimization and Congestion Control in Datacenters. In both cases, MAGNNETO has achieved similar performance to state-of-the-art approaches, offering significant improvements in execution cost while being totally compliant with current and legacy equipment.

In this talk, we will explore the MAGNNETO framework in detail, discussing its innovative design, key features, and multifaceted applications. In addition to this, we will also introduce a promising research line to enhance the framework by integrating Topological Deep Learning techniques, envisioning a novel approach that goes beyond the traditional graph domain.

Join us as we embark on a pioneering journey to redefine the landscape of distributed network optimization and management with MAGNNETO .

Link to the recorded talk: https://drive.google.com/file/d/1bDpFDyEoLNOwRg4CU5UK6yVaMqmwq1rw/view?usp=sharing

This talk is part of the Foundation AI series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Artificial Intelligence Research Group Talks (Computer Laboratory)
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge Forum of Science and Humanities
• Cambridge Language Sciences
• Cambridge talks
• Chris Davis' list
• Department of Computer Science and Technology talks and seminars
• Guy Emerson's list
• Hanchen DaDaDash
• Interested Talks
• Lecture Theatre 2, Computer Laboratory, William Gates Building
• School of Technology
• Speech Seminars
• Trust & Technology Initiative - interesting events
• bld31
• ndk22's list
• ob366-ai4er
• rp587
• yk373's list

Note that ex-directory lists are not shown.