Online Event from 21.-25.2.2022 and In-person Event from 18.-22.07.2022

The Summer School

Communication networks have seen significant advances in recent years including trends such as software-defined networking, programmable data planes, network hardware acceleration and network function virtualization. The aim of this summer school is to introduce PhD students to currently relevant topics in the intersection of programmable networks and AI/machine learning. The first part of the summer school is a mini-course consisting of a series of 90-minute lectures that will be given online in the week from 21.02.2022 until 25.02.2022. In addition to the online mini-course a physical meeting is planned to take place in-person in the beautiful city of Ulm in southern Germany from 18.07.2022 -22.07.2022.

The mini-course will start with an introduction to the state of the art in AI and machine learning and then include topics like decision making in network management, network and distributed systems programming and boosting network applications using machine learning techniques. Other topics include traffic prediction and classification, routing and congestion control, network resource management and QoS provisioning as well as applications in the area of network security.

During the in-person part of the summer school, participants will be given the opportunity to introduce their own researchin short lightning talks but the majority of time will be reserved for discussions and working groups. Between the two parts of the summer school, we will provide a slack workspace to allow discussions to continue online.

Summer School Organizers

Amr Rizk, University of Duisburg-Essen
Frank Kargl, Ulm University
Olaf Landsiedel, University of Kiel

Event Organization

Online Event 21.-25.02.2022

Lecture-length Talks by experts (see list below): Each talk 90min talk + 15min Q&A
Lectures will be spread over 5 days with a load of 90-180 min per day.
Lecture times will be set in preference to the speaker time zones and will be announced as soon as possible!

In-Person Event 18.-22.07.2022

A call for lightning talks will be sent to summer school participants during the week of the online event. Planning for working groups will also start then depending on participants interests. The in-person event takes place at Schloss Reisensburg in Ulm and will also include time for social events.

Call for Participation

The summer school AI in Networking focuses on approaches to address problems and challenges in networking through methods from AI/ML. Examples of such application areas include network management techniques, congestion control, resource optimization, and network security..

The summer school takes place in two parts: The first part is online from 21.02.2022 until 25.02.2022 and the second part is in-person in Ulm, Germany from 18.07.2022 -22.07.2022. The number of attendees will be limited to max. 20 persons. Registrations are open from Dec. 15. Participation to the online part will be open to everyone. If participation to the in-person part has to be restricted, the organizers will select participants based on the extended abstracts.

We encourage in particular PhD students and early career researchers to are welcome to participate in the summer school, especially when working on network management techniques, congestion control, resource optimization, and network security or in a related field related to the topic of the summer school. Master students and post-docs are are also welcome to participate upon if availability allows.

Concept of the Summer School

The aim of the summer school is to spark joint work between the participating PhD students within the area of AI/ML for networking while benefiting from input and discussions with senior researchers in the field. To achieve this, the online mini-course will serve as a common introduction into different aspects of the topic of the summer school. During the time between the online part and the in-person part of the summer school, participants are given the opportunity to self-organize into interest groups over a Slack workspace. During the in-person part of the summer school, lightning talks and posters allow to present own research work, working groups will allow to discuss and develop their ideas and initiate new collaboration. We will conclude this week with a mini-poster session where results will be presented.

Participants are strongly encouraged to submit an extended abstract of at most two pages that describes their current research (ideas, interests, or simply motivation) in relation to the topic of the summer school. Participants will be given opportunities to present their work in lightning talks and poster sessions during the presence week. The extended abstracts must be formatted according to the ACM conference proceedings format.

Important Dates

Abstract registration: until 01.02.2022

Notification of acceptance: 11.02.2022

Online Event: 21.-25.02.2022

In-Person Event: 18.-22.07.2022

Registration

  • Application for the summer school opens on 15.12.2022 and closes 01.02.2022.
  • To register for the summer school send your extended abstract (at most two pages in ACM format) by 01.02.2022.
  • Please submit your extended abstract here
  • Confirmations of the registration will be sent to the participants by 11.02.2022
  • Every registrant will be able to participate in online talks, if participation to the in-person part needs to be restricted, selection will be based on the submitted abstracts.
  • The summer school will not charge for the attendance, however, participants are expetced to pay their own travel expenses in addition to a fee for the room and board at Schloss Reisenburg which is paid directly at the venue. Preliminary figures are in the range of 100-120 EUR per night incl. full board. The exact figures will be published very soon.

Program

Confirmed speakers

Program - Online event 21.-25.02.2022 - all times are CET

Monday 21.02.2022

08:15-09:00  •  Welcome & icebreaker (introduction round)
09:00-12:00  •  Mini course  •  Vasileios Belagiannis
                                                Machine learning and deep learning basics

Tuesday 22.02.2022

09:15-10:45  •  Mini course  •  Stefan Schmid
                                                Principles of self-* communication networks: data-driven optimization and verification
10:45-11:15  •  Coffee break & breakout rooms
11:15-12:45  •  Mini course  •  Tanja Zseby
                                                Machine Learning for Network Security (Abstract)

Wednesday 23.02.2022

11:00-12:30  •  Mini course  •  Carsten Binnig
                                                Learned DBMS Components 2.0: From Workload-Driven to Zero-Shot Learning (Abstract)
12:30-13:00  •  Coffee break & breakout rooms
14:00-15:30  •  Mini course  •  Aaron Ding
                                                Bring Intelligence to the Edge: Lessons and Roadmap for Edge AI (Abstract)

Thursday 24.02.2022

09:15-10:45  •  Mini course  •  Andreas Blenk
                                                Need for Speed: Applying Machine Learning to Network Management
10:45-11:15  •  Coffee break & breakout rooms
11:15-12:45  •  Mini course  •  Fabien Geyer
                                                Performance Evaluations of Computer Networks with Graph Neural Networks (Abstract)

Friday 25.02.2022

09:15-10:45  •  Mini course  •  Nic Lane
                                                Federated Learning 101 and Research Opportunities using Flower
10:45-11:15  •  Coffee break & breakout rooms
14:00-15:30  •  Mini course  •  Konrad Rieck
                                                Not that smart! Security Vulnerabilities in Machine Learning (Abstract)
15:30-15:45  •  Closing and preliminary information on the in-person event

In-Person Event

The In-Person event will take place 18.07.2022 - 22.07.2022 Schloss Reisensburg, University of Ulm, Günzburg, Germany.
Tentative program for the In-Person event:

Time Monday 18.07. Tuesday 19.07. Wednesday 20.07. Thursday 21.07. Friday 22.07.
   
8:00 - 9:00   
   
Arrival   
   
Breakfast   
   
Breakfast   
   
Breakfast   
   
Breakfast   
   
9:00 - 10:30   
   
Arrial   
   
Talk by A. Blenk   
   
Tutorial Adv. Examples   
   
Hackathon Opening   

Presentation of
Hackathon results
   
10:30 - 11:00   
   
Arrival   
   
Break   
   
Break   
   
Break   
   
Break   
   
11:00 - 12:30   
   
Arrival   
   
Shadow TPC Session 1   
   
Tutorial   
   
Hackathon Worktime   
Presentation of
Hackathon results
& Closing
   
12:30 - 13:30   
   
Lunch   
   
Lunch   
   
Excursion Legoland   
   
Lunch   
   
Lunch   
   
13:30 - 15:00   
   
Poster Session 1   
   
Shadow TPC Session 2   
   
Excursion Legoland   
   
Hackathon Worktime   
   
Return travel   
   
15:00 - 15:30   
   
Break   
   
Break   
   
Excursion Legoland   
   
Break   
   
Return travel   
   
15:30 - 17:00   
   
Poster Session 2   
   
Shadow TPC Session 3a   
   
Excursion Legoland   
   
Hackathon Worktime   
   
Return travel   
   
17:00 - 18:30   
   
Break   
   
Shadow TPC Session 3b   
   
Excursion Legoland   
   
Break   
   
Return travel   
   
18:30 - 19:30   
   
Dinner   
   
Dinner   
   
Barbecue   
   
Dinner   
   
Return travel   
19:30 - Socializing Socializing Socializing Socializing

Travel
You can reach Günzburg via the German railway (stop Günzburg).
On this map, you see markings the train station, the Castle Reisensburg, and Legoland.
To get from the train station to Reisensburg, there are 3 Options:

  • Use the bus. Get out at stop "Reisensburg - Mitte“
  • Walk the 2.2 km from the train station
  • Use a taxi

Arrival on Sunday
For those traveling from afar, it might not be possible to arrive on Monday morning. In this case, you can already checkin on Sunday.
This should ideally be between 17:00 - 18:00.
If you want to use this option, please contact the Reisensburg staff at the email address you received as soon as possible to arrange your extra night.

Hygiene Concept
Currently, Reisensburg regulations foresee a number of COVID-related measures. The most important ones are:

  • There is a 3G requirement, i.e., you need to either provide a proof of vaccination, proof of COVID-recovery, or a certified test on checkin.
  • Mask wearing is mandatory while moving in the public areas. Exception is the seminar room. Of course we hope for good weather to allow us to also spend breaks etc. outside.
  • Contact persons of COVID-positive patients or persons showing COVID-related symptoms are not allowed to participate.
The full regulations will be sent to all attendees via email.

Hackathon Preparation
We want to run a hackathon with you, where you will work in small groups of topics of your choice (or course related to AI & Networking). For this and to form groups, you should propose topics to us that you always wanted to work on (but never found the time) and that should be doable in a day of coding.
In order to see if we will have enough topics, please send us your proposals by 25.06.2022.
A proposal should include a rough description of a topic and maybe some information on resources to be used (datasets, ML frameworks, what resources do you need..)

What to bring to the In-Person Event?
Make sure to bring a laptop for the practical sessions and don’t forget to also bring a A0 (portrait) poster for the poster session to present and discuss your research.

Important Dates

Submission Deadline for Extended Abstract (max. 2 pages): 01.02.2022

Notification of participation:
08.02.2022

Summer School date (Online part):
21.02.2022 - 25.02.2022
via Zoom

Summer School date (In-person part):
18.07.2022 - 22.07.2022
Schloss Reisensburg, University of Ulm, Germany

Sponsors
This Summer School is powered by MAKI
Collaborative Research Centre 1053 -
Multi-Mechanisms Adaptation for the Future Internet

MAKI