Monday February 18, 2019 - AIMLEM Program

14:00 - 14:10  Welcome and Introduction

14:10 - 15:30   Technical Session 1

1. An Approach to Apply Reinforcement Learning for a VNF Scaling Problem
   Hai Nguyen and Tien Van Do (Budapest University of Technology and Economics, Hungary);Attila Hegyi (Nokia - Bell Labs Research, Hungary); Csaba Rotter (Nokia - Bell Labs Research, Hungary)

2. Improving the QoE of DASH over SDN: A MCDM Method with an Intelligent Approach
   Simge Özcan and Muge Sayit (Ege University, Turkey)

3. Toward Real-time Packet Classification for Preventing Malicious Traffic by Machine Learning
   Toki Suga (The University of Tokyo, Japan); Kazuya Okada (The University of Tokyo & ATR, Japan);Hiroshi Esaki (The University of Tokyo, Japan)

15:30 - 16:00   Coffee Break

16:00 - 17:00   Keynote talk + Q & A session

17:00 - 17:50   Technical Session 2

1. WCDMA Mean User Throughput Prediction Using Linear Regression Algorithm
   Mourad Nasri (Higher School of Communicatins of Tunis, Carthage University, Tunisia);Mohamed Hamdi (University of Carthage, Tunisia)

2. LTE QoS Parameters Prediction Using Multivariate Linear Regression Algorithm
   Mourad Nasri (Higher School of Communicatins of Tunis, Carthage University, Tunisia);Mohamed Hamdi (University of Carthage, Tunisia)

17:50 - 18:00   Wrap up

AIMLEM 2019 Registration

AIMLEM 2019 is a half-day conference, co-located with ICIN 2019. Check AIMLEM 2019 for details.

Registration can be done solely to AIMLEM 2019, but a reduced attractive rate is available for a full event including ICIN 2019.

We encourage AIMLEM 2019 participants to register the full event.

AIMLEM 2019 Keynote

Elena Fersman (Ericsson, Sweden)

Title: Intelligent Network Operations with Artificial Intelligence

Abstract: This talk will cover several aspects of the network operations of the future, done in an automated, proactive, and intent-driven way. I will present different use cases based on digital twins of telecom sites learnt over time and mechanisms for decision-making based on those in a distributed real-time infrastructure.

Biography: Elena Fersman is a Research Director in Artificial Intelligence at Ericsson. She is responsible of a team of 100+ researchers located in Sweden, US, India, Hungary and Brazil. She is also a docent and an adjunct professor in Cyber-Physical Systems specialized in Automation at the Royal Institute of Technology in Stockholm. She holds a PhD in computer science from Uppsala University, a Master in Management from St. Petersburg Polytechnic University and did a postdoc at École Normale Supérieure Paris-Saclay. At Ericsson, she had various positions ranging from product management to research leadership. Her current research interests include automation of knowledge-intensive cyber-physical systems. Elena has co-authored over 20 scientific publications and over 50 patent families.

1st International Workshop on Artificial Intelligence and Machine Learning Techniques for Enhanced Network Management
(AIMLEM 2019)

(Click here to download CFP in PDF format)

The current interest in both Artificial Intelligence (AI) and Machine Learning (ML) techniques in the context of network management has come about due to the requirement to address the complex management of software defined infrastructures, including slices, SDN, NFV, and SFC features, which are beyond the reasonable input, scope and ability for direct human interaction. Progress in both the performance of CPUs as well and the performance and accuracy of machine learning methods such as neural networks, has made Artificial Intelligence and Machine Learning realistic approaches for use in network management.

The AIMLEM workshop addresses both the advances and challenges related to Artificial Intelligence and Machine Learning techniques for enhanced network management of network elements and services in current and future highly dynamic and highly scalable 5G environments.

The advances and challenges are expected to be multiple, and there are clearly many open questions that need to be addressed, including:

• How can Artificial Intelligence and Machine Learning Techniques really be used for effective and/or enhanced network management;


• What are the abstractions and knowledge representation / data models needed to ensure that AI is deployable for network management and orchestration;


• How do the existing technologies of networking, NFV, SDN, services become features and aspects of AI and ML, and how are they managed in this context;


• Is it better to adapt existing components to support AL and ML, or is it better to design new ones, considering the price / performance trade-offs for introducing AI/ML in management and orchestration;


• How can the use of induction and explanation systems of AI highlight what decisions have been made and how these situations have occurred;

AIMLEM aims at providing an international forum for researchers and practitioners from academia, industry, network operators, and service providers to discuss and address the challenges deriving from such emerging scenarios where AI and ML systems, processes, and workflows used in both service and network domains. The workshop welcomes contributions from both computing and network-oriented research communities, with the aim of facilitating discussion, cross-fertilization and exchange of ideas and practices, and successfully promote innovative solutions toward a real use of AI and ML. Contributions that discuss lessons learnt and best practices, describe practical AI and ML deployment and implementation experiences, and demonstrate innovative AI and ML use-cases are especially encouraged for presentation and publication.

We are interested in papers that use Artificial Intelligence and/or Machine Learning the following topics:

• Distributed versus centralised management architectures and algorithmic approaches of AI and ML for 5G


• Dynamic management and orchestration for virtualized features of NFV, SDN, and SFC (including function placement, network slicing)


• AI assisted network and cloud slicing


• Data, information, and semantic models, and abstractions and knowledge representation for AI systems in network management


• Network data / metadata collection, analysis, distribution, and visualisation for operations and testing of AI/ML methods and algorithms


• Evaluation (performance and feasibility) of integrating AI and ML into the networks (e.g., operation, management and orchestration)


• Success scenarios of AI/ML demonstrated in network management


• And other AI / ML management and orchestration related topics

Submission guidelines

Submitted papers must be original work, not under review at other journals/conferences, and may comprise a maximum of 6 A4 (210 mm x 297 mm) pages in 2-column IEEE conference style with a minimum font size of 10 pt. Papers should be submitted electronically using the EDAS online submission system. All accepted papers must be presented by one of the authors.
Submission: http://edas.info/N25418

Proceedings

Papers accepted for AIMLEM 2019 will be included in the conference proceedings and IEEE Xplore. The IEEE reserves the right to remove any paper from IEEE Xplore if the paper is not presented at the workshop.

Workshops Co-Chairs

• Stuart Clayman (University College London, UK)
• Slawomir Kuklinski (Orange Labs, Warsaw, Poland)
• Qiong Zhang (Fujitsu Laboratories, Dallas USA)
• Min Xie (Telenor Research, Norway)
• Rashid Mijumbi (Nokia Bell Labs, Dublin, Ireland)