Abstract: Network Intelligence course encompasses three main parts. In the first part of the course, we aim to present the background on Network Intelligence i.e. the latest state of the art on Artificial Intelligence (AI) for network and service management. In the second part of the course, we will delve into Machine Learning (ML) for network data analysis. We will be introducing and discussing various techniques for specific problems (prediction, classification, clustering, etc.), as well as relevant pointers to consider in the domain. In the third part, we will present a hands-on demonstration through a notebook capturing the end-to-end process of leveraging ML in a network management use case (forecasting of network data, anomaly detection, etc.). The demonstration will make use of real network data. The participants will be presented with:
- An overview of past literature works and ongoing efforts on applying ML for network operations and management.
- A concise course on applying ML for network data analytics with relevant pointers and useful takeaways.
- Insights about best environments to use (opensource libraries and frameworks)
- A hands-on demonstration showcasing ML techniques applied to a network management use case scenario that involves real network traces.
Teaching and hands-on experience of the instructors will be leveraged for best possible outcome.
Imen Grida Ben Yahia is currently with Orange Labs, France, as a Research Project Leader on Autonomic & Cognitive Management and Expert in Future Networks. She is also leading the international initiative within IEEE comsoc on “Network Intelligence” http://ni.committees.comsoc.org/. She received her PhD degree in Telecommunication Networks from Pierre et Marie Curie University in conjunction with Télécom SudParis in 2008. Her current research interests are autonomic and cognitive management for software and programmable networks that include artificial intelligence for SLA and fault management, knowledge and abstraction for management operations, intent- and policy-based management. She contributed to several European research projects like Servery, FP7 UniverSelf, the H2020 CogNet and currently the 5G SliceNet. Imen is teaching Cognitive Network Management in Telecom Sud Paris, a module of (6 to 12) hours per year including Network management operations challenges, use case for Cognitive Network Management (AI based) and hands-on showcasing machine learning for real network data.
Noura Limam received the M.Sc. and Ph.D. degrees in computer science from the University Pierre et Marie Curie, Paris VI, in 2002 and 2007, respectively. She is currently a research assistant professor of computer science at the University of Waterloo, Canada, where she has been teaching 400- and 600-level courses in computer networks and distributed systems (CS456/CS656, CS454/CS654, and CS436/CS636) with combined theory and hands-on training, for the last several years. She is on the technical program committee and organization committee of several IEEE conferences. Her contributions are in the area of network and service management. Her current research interests are in autonomic networking and cognitive network management.
Communication networks have evolved drastically in the last decade. Whereas networks largely used to provide dumb connectivity pipes interconnecting its end users, current network technology is tightly interconnected with the cloud, leading to a plethora of advanced services, a drastic increase in network usage, and strongly evolved data and control planes. Software-based functionality is now deeply changing the nature of both the control and data plane of our networking infrastructure through SDN and NFV technology respectively. This has introduced tremendous programmability and flexibility but also a range of uncertainties in the performance, security and management of our networks. Less functionality is now specified in standardized protocols or hardcoded in our data plane hardware.
Up to recently, machine learning techniques have been used in networking mostly for mechanisms outside of the control loop and outside of the fast path of our networks. Whereas machine learning has been effectively used for anomaly detection, network prediction or analytical purposes, the increasing network softwarization is now creating potential for apply recent evolutions in machine learning to optimize the actual control- as well as the data plane operation of networking infrastructure. Such applications might, for example, learn and optimize network performance relationships between softwarized network functions, presented loads and potential software/hardware configurations. The increasing availability of monitoring data and associated monitoring platforms might even further fuel advanced ML techniques including deep learning in exploiting ML-driven network approaches.
This workshop aims to bring together network researchers and machine learning experts with network operators and vendors to identify key opportunities, challenges and preliminary solutions for machine learning in softwarized networks. This includes the discussion of relevant data sets, best practices, and/or insights from earlier attempts. Particular topics of interest, but not limited to the following are:
- ML for performance analysis & optimization of (softwarized) networks
- ML for performance optimization of (softwarized) networks
- Security improvement through the use of Machine Learning
- Network Data Plane and HW acceleration techniques in support of ML
- ML for closed-loop network control (e.g., network resource allocation in a wireless network, computing MAC schedules or taking handover decisions)
- ML in Software-Defined Networking
- Network orchestration enhanced with ML (e.g., scaling, placement, admission control, load-dependent flow routing, …)
- Monitoring techniques in support of network ML (e.g., compressing monitoring data to meaningful results or deciding where in a network monitoring is required to detect inconsistent behavior)
This workshop is a satellite workshop of the NetSoft 2020 conference June 29-July 3, 2020, in Ghent, Belgium.
Interested authors are invited to submit papers of up to 8 pages (including references), presenting industrial innovations, work in progress, research results or technical developments.
Accepted and presented workshop papers will be published in the conference proceedings and will be submitted to IEEE Xplore. For more details about submission, please check the following websites:
- Submission link for NetLearn 2020: EDAS link
- Further submission information: https://netsoft2020.ieee-netsoft.org/authors/
Important: Please check NetSoft 2020 publication and no-show policy in the conference website at https://netsoft2020.ieee-netsoft.org/authors/publication-and-no-show-policy/.
- Paper submission: February 14th, 2020
- Notification of acceptance: March 23rd, 2020
- Camera-ready paper: April 6th, 2020
- Wouter Tavernier, Ghent University & imec, Belgium
- Holger Karl, Paderborn University, Germany
- Alex Galis, University College of London, United Kingdom
- Antonio Manzalini, Telecom Italia, Italy
- Arunselvan Ramaswamy, Paderborn University, Germany
- Ahmed M. Abdelmoniem, KAUST, Saudi Arabia
- Bin Cheng, NEC, Germany
- Chadi Barakat, INRIA, France
- Christian Rothenberg, University of Campinas, Brazil
- Feng Li, Shandong University, China
- Fulvio Risso, Politecnico di Torino, Italy
- Gianluca Reali, CNIT, Italy
- Mats Bengtsson, KTH, Sweden
- Olaf Landsiedl, Kiel University, Germany
- Oliver Hohlfeld, Brandenburg University of Technology, Germany
- Pan Hui, University of Helsinki, Finlan
- Philippe Owezarski, LAAS-CNRS, France
- Rebecca Steinert, RISE, Sweden
- Rolf Stadler, KTH, Sweden
- Stefan Schmid, University of Vienna, Austria
- Wolfgang John, Ericsson, Sweden