User Tools

Site Tools


home

KuVS Fachgespräche: Machine Learning & Networking

This workshop is associated with the ERC Consolidator Grant Project FlexNets “Quantifying Flexibility in Communication Networks” that has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program grant agreement No 647158 – FlexNets (2015 – 2020). http://www.networkflexibility.org

Topic: Machine learning in the context of communication networks
Begins: February 20th, 2020, 11:45
Ends: February 21st, 2020, 13:00
Venue: Arcisstr. 21, 80333 Munich, Germany, Room 1977
How to find the room? https://portal.mytum.de/campus/roomfinder/roomfinder_viewmap?mapid=12&roomid=1977@0509
Fee: Free of charge (social dinner is self paid)
Organizers:TUM, UPB, KIT
Contact: fgmln2020@easychair.org

Machine learning and artificial intelligence (ML/AI), in particular deep learning, has led to breakthroughs in various domains such as image recognition or natural language processing.This workshop focuses on the topic of ML/AI in the context of communication networks. It aims to discuss research visions and results as well as opportunities and challenges in the intersection of these two areas. The workshop looks for contributions and ideas that provide useful combinations of ML/AI approaches to address networking challenges on all layers from MAC to Application.

Agenda Day 1 (Thu 20.02.2020)

Start End Event
11:45 12:00 Small snacks
12:00 12:15 Welcome by Andreas Blenk and Prof. Kellerer
12:15 13:15 Session 1: Can ML finally solve congestion control problems?
Session chair: Andreas Blenk (TUM)

Talk 1: Using Deep Learning in Network Measurements for Passive Congestion Control Identification; C. Sander, J. Rüth (RWTH Aachen), O. Hohlfeld, K. Wehrle (BTU)

Talk 2: TCP Congestion Control Using Imitation Learning; B. Jaeger, J. Schmeißer (TUM)
13:15 13:45 Teaser Session for Posters and Demos

* NOracle: Who is communicating with whom in my network? (TUM-LKN)
* The Softwarised Data Zoo (UPB)
* Learning from Hierarchical Heavy Hitters (KIT)
* Artificial Intelligence and Machine Learning at LKN (TUM-LKN)
* FlexNets: Quantifiying Flexibility in Communication Networks (TUM-LKN)
* Veni Vidi Dixi: reliable wireless communication with depth images (TUM-LKN)
* NCSbench: Reproducible Benchmarking Platform for NCS (TUM-LKN)
13:45 14:15 Short Break with Posters and Demos
14:15 15:45 Session 2: ML for Network Modeling
Session chair: Stefan Schneider (UPB)

Talk 1: DeepMPLS: Fast Analysis of MPLS Configurations Using Deep Learning; F. Geyer (TUM)

Talk 2: Runtime Verification of P4 Switches with Reinforcement Learning; A. Shukla, K. N. Hudemann (TU Berlin), A. Hecker (Huawei), S. Schmid (University of Vienna)

Talk 3: Optimising the Performance of Deep Transfer Learning for Communication Networking Applications; T. V. Phan, T. Bauschert (TU Chemnitz)
15:45 16:30 Group photo & Longer Break & 5G Demo
16:30 18:00 Session 3: Reinforcement Learning for Wireless
Session chair: Murat Gürsu (TUM)

Talk 1: Reinforcement Learning Scheduler for V2V Communications under Intermittent Coverage; T. Sahin, M. Boban, R. Khalili (Huawei), A. Wolisz (TU Berlin)

Talk 2: CBMoS: Combinatorial Bandit Learning for Mode Selection and Resource Allocation in D2D Systems; A. Ortiz, A. Asadi, M. Engelhardt, A. Klein, M. Hollick (TU Darmstadt)

Talk 3: Towards Delay-Minimal Scheduling through Reinforcement Learning in IEEE 802.15.4 DSME; F. Meyer, V. Turau (TU Hamburg)
19:30 open Social dinner (self-paid) at Park Café München

Agenda Day 2 (Fri 21.02.2020)

Start Stop Event
08:30 09:00 Small snacks
09:00 10:30 Session 4: ML against Privacy
Session chair: Hauke Heseding (KIT)

Talk 1: Machine Learning-based Real-time Estimation of Quality of Experience from Encrypted Video Streaming Traffic; N. Wehner, M. Seufert (University of Würzburg), S. Wassermann, P. Casas (AIT), T. Hoßfeld (University of Würzburg)

Talk 2: Reducing Consumed Data Volume in Bandwidth Measurements via a Machine Learning Approach; C. Maier, P. Dorfinger, J. L. Du (Salzburg Research Forschungsgesellschaft), S. Gschweitl, J. Lusak (alladin-IT)

Talk 3: A Concept for Crowd-sensed Prediction of Mobile Network Connectivity; S. Herrnleben, B. Zeidler, M. Züfle, C. Krupitzer, S. Kounev (University of Würzburg)
10:30 11:00 Short break with Posters and Demos
11:00 12:00 Session 5: Prediction and Incremental Learning
Session chair: Robert Bauer (KIT)

Talk 1: Prediction for Location-Aware Network Automation in Radio; M. Kajó (TUM), J. Ali-Toppa (Nokia Bell Labs)

Talk 2: Deployment of Incremental Learning Methods for Neural Networks in Resource Allocation Problems; L. Fisser, S. Linder, A. Timm-Giel (TU Hamburg)
12:00 13:00 Closing & Lunch & Chill-out

Deadlines

Submission January 17th, 2020 https://easychair.org/conferences/?conf=fgmln2020
Notification January 24th, 2020 Notifications will be sent per mail
Final submission & Registration January 31st, 2020 Details for registration will follow

Submission Guidelines

  • Extended abstracts of up to 2 pages
  • PDF format, formatted according to the double-column IEEE format
  • The workshop is open to submissions containing preliminary and previously published work

Topics of Interest

Topics of interest include but are not limited to:

  • Data mining & visualization, statistical modeling, and big data analytics for networking data
  • Frameworks or tools for data analytics or visualization for networking data
  • Time series predictions for networking data such as traffic demands, failures, etc.
  • AI/ML algorithms for anomaly detection and attack step prediction in network security
  • Protocol design and optimization using ML/AI
  • Deep learning and reinforcement learning in network control & management
  • Resource allocation for virtualized networks using machine learning
  • Machine learning & transfer learning for prediction of networking data & control decisions
  • Practical implementations or experience with ML/AI in networking
  • Self-learning and adaptive networking protocols and algorithms
  • Self-X networks: Self-learning, self-driving, self-repairing, etc.
  • New concepts like empowerment for quantifying and improving ML/AI-based concepts

Organizers

TPC

  • Oliver Hohlfeld, BTU
  • Frank Kargl, Ulm University
  • Holger Karl, Paderborn University
  • Wolfgang Kellerer, TUM
  • Michael Menth, University of Tübingen
  • Amr Rizk, Ulm University
  • Stefan Schmid, University of Vienna
  • Oliver Waldhorst, Hochschule Karlsruhe
  • Martina Zitterbart, KIT
home.txt · Last modified: 2020/02/20 11:45 by guest