Logo

ITU AI/ML in 5G Challenge Grand Challenge Finale (Final Conference) - Shared screen with speaker view
Albert Cabellos
06:15:12
Hi all
NARANMANDAKH Tumen-Ulzii
06:15:37
Hello everyone, Dr Naranmandakh T, Greetings from the Communications Regulatory Commission of Mongolia,
Vishnu Ram
06:16:31
Welcome :)
Yoichi MAEDA
06:17:08
Hellow Vishnue;
Reinhard Scholl
06:19:04
Reinhard Geneva
Jay Patel
06:19:06
Hello everyone from Nairobi, Kenya
ITU AI for Good
06:19:09
Geneva, Switzerland
Pramod Misra
06:19:13
Mumbai,India
Rewant Prakash
06:19:13
New Delhi, India
Brigitte BRUN
06:19:13
Kind regards from Germany
Abhishek Dandekar (TU Berlin)
06:19:15
Berlin, Germany
Aldebaro Klautau
06:19:15
Hello from Brazil, Belem
Miquel Ferriol Galmés
06:19:16
Barcelona, Spain
Yoichi MAEDA
06:19:19
Tokyo Japan
Vishnu Ram
06:19:21
Bangalore, India
François Schnitzler
06:19:22
Vannes, France
Bahare Masood Khorsandi
06:19:22
Munich Germany
Francisco Muller
06:19:23
Belem, Brazil
Virendra Singh Rajput
06:19:23
Bangalore, India
Eoghan Furey
06:19:24
Dia Duit (Hello) from Donegal, Ireland :)
Charles Ssengonzi
06:19:25
Johannesburg South Africa
Özlem Tugfe Demir
06:19:25
Hello everyone! Linköping, Sweden
Ahmad Nagib
06:19:25
Kingston, Canada
Szymon Kobus
06:19:26
London, UK
Muneaki Goto
06:19:27
Hi, attending from Tokyo, Japan:-)
Ramon Vallés Puig
06:19:29
Barcelona, Spain
Atheer Alsaif
06:19:31
Riyadh, Saudi Arabia
Scott (Ericsson)
06:19:34
Pittsburgh, Pennsylvania USA
CARMELO JOSE ALBANEZ BASTOS FILHO
06:19:38
Recife, Brazil
Shagufta Henna
06:19:46
Shagufta, LYIT, Ireland
Majid Naderkhani
06:19:46
Maryland USA
Yuusuke Hashimoto
06:19:59
Hi, attending from Osaka, Japan
Karim Rabie
06:20:02
Hello from Cairo, Egypt.
Alejandro Alba
06:20:05
Geneva
Mohammad Abid
06:20:06
Mohammad Abid, stc, Riyadh, Saudi Arabia
Ryuma Kinjo
06:20:09
Okinawa, Japan
Armen Aghasaryan
06:20:23
Paris
Willy Fitra Hendria
06:20:27
Willy from Indonesia
MP SINGHAL
06:20:28
Greeting to all -MP Singhal New Delhi IndiIa
Cheng Qiang
06:20:29
Beijing, China
José Suárez-Varela
06:20:32
Barcelona, Spain
Nao Uehara_HENOKOKING
06:20:37
Hello from Okinawa Japan.
Regina Valiullina
06:20:59
Geneva, Switzerland ;)
gang zhouwei
06:21:40
Guizhou, China. :)
Mohammad Malekzadeh
06:22:37
London
qian deng
06:23:41
china
Arzu Alpagut
06:24:04
Greetings form Istanbul
Artem Volkov
06:24:09
St.-Petersburg, Russia
Thomas Basikolo
06:27:25
https://www.itu.int/en/ITU-T/AI/challenge/2020/Pages/PROGRAMME.aspx
Ammar Muthanna
06:37:00
With which product did you compare yours ? do have any results?
Akihiro Nakao
06:41:02
I have a follow on question to vishunu
Ahmad Nagib
06:58:57
Do you have any justification why MLP is giving you the worst performance among all the predictors used? could it be due to manual feature refinement and reduction carried out?
Mengfei Feng
07:00:40
why Tree model work better than MLP and SVM in this problem?
qian deng
07:04:43
p0[
Ahmad Nagib
07:14:55
Any reason why the recall of your trained model is low?
Ahmad Nagib
07:16:03
the GCN model specifically
Xing Wang
07:19:26
why GCN performs better than other models, can you explain based on the problem
Hesham Elbakoury
07:43:17
how we can get the slides of this event.
Thomas Basikolo
07:44:25
The slides and information can be found on the website
Ahmad Nagib
07:56:16
What kind of environment did you use to test your solution?.. simulation?.. real network deployment?
Ahmad Nagib
07:59:29
Also when you say traffic prediction.. do you mean the overall traffic on the network?
Ahmad Nagib
08:14:09
What is the type of nodes in your scenario here (in the context of 5G)? and is it practically easy to introduce these changes to the topology?
Mohammad sharique Ahmad
08:15:02
that was wonderful , for 44.5% improvement simulation , what was the duration of data what was taken into consideration?
ravi D
08:39:51
questions: 1. how much % energy saving will be achieved from this approach?. 2. any sensor input considered as data source?. 3. what all data inputs expected for this approach?
Cresting - Shiyi Zhu
08:53:40
answer for question 1:We don't know how much energy saving because we just have the data of everyday's energy-saving status.
Cresting - Shiyi Zhu
08:56:16
answer for question 2:We have no access to real-time data source. If sensor input is offline, we should have considered.
Cresting - Shiyi Zhu
08:57:35
answer for question 3:Historical base station performance data, historical energy saving status and base station configuration information.
Vishnu Ram
08:59:38
Mr. Ming: is there a relation with standards?
Paola Soto - ATARI
09:11:01
What kind of algorithms are used for the orchestration?
Abhishek Dandekar
09:18:38
@Thomas when will we know results of public voting
Thomas Basikolo
09:20:35
The results will be announced on Thursday
Abhishek Dandekar
09:20:52
thanks
Ahmad Nagib
09:26:22
Did you consider evaluating the cost of the communication overhead added by applying federated learning?and also from the aspect of practicality, will these distributed devices be able to carry out the local training tasks? as in most of the federated learning cases the training task is simpler than the one in your work!
Ahmad Nagib
09:31:41
no problem!
Vishnu Ram
09:32:19
Thank you Ahmad
Mahdi Boloursaz Mashhadi
09:34:05
For NU Huskies: Do U have results for infusing two out of three of the modalities (image, lidar, coordinates)? How would a network infusing only LIDAR and coordinate data work? How can you be sure that your proposed detection and background removal on image data is actually the reason for improvements  in the final result when 3 modalities are infused? Do U have an ablation study to show contribution of your proposed detection and background removal on the final performance given the fact that it increases the complexity of NN significantly? 
Thomas Basikolo (ITU)
09:39:02
For Participants in The ITU AI/ML in 5G Challenge, please respond to this survey: https://forms.office.com/Pages/ResponsePage.aspx?id=12TkI-YEh0uRPCS9iSGf0-yqkfLCoQ9IpTbc_XELf95URExJWlM2OExXVEQ5VFhFSTUzWUU3SDU1WS4u
Bo Lv-CAICT
09:39:13
maybe some training packet could be dropped to reduce communication overhead because ML training load distribution has long tail. Removing some redundant training data may not impact the final performance of ML training.
Thomas Basikolo (ITU)
09:41:03
The Grand Challenge Finale (Keynotes + Awards) will take place on 17 Dec. Please register here: https://itu.zoom.us/webinar/register/WN_bm4oJ_OeRH-Ilnb4GY_gMQ
Mahdi Boloursaz Mashhadi
09:41:18
For NU Huskies: Do U have results for infusing two out of three of the modalities (image, lidar, coordinates)? How would a network infusing only LIDAR and coordinate data work? How can you be sure that your proposed detection and background removal on image data is actually the reason for improvements  in the final result when 3 modalities are infused? Do U have an ablation study to show contribution of your proposed detection and background removal on the final performance given the fact that it increases the complexity of NN significantly? 
Bo Lv-CAICT
09:50:48
If fc layer can be replaced by other opertaion such as average pool or 1*1 conv to reduce complexity of NN ? because fc(fully connection layer) may have so many parameters. My opinion..
abdi shakib
09:52:53
When is the next challenge?
Thomas Basikolo
09:54:07
Please tune-in (join) in 17 Dec to get information about what happens next: https://itu.zoom.us/webinar/register/WN_bm4oJ_OeRH-Ilnb4GY_gMQ
Mikolaj Jankowski
09:54:38
For BeamSoup: What is beta in your loss function? Did you analyse the impact of beta by some ablation? It seems like such loss should force your outputs to two different distributions (one hot encoded best beam vs. signal power), can you please elaborate on that a bit?
Mahdi Boloursaz Mashhadi
09:55:13
Beam soup: The way the baseline pre-process LIDAR data at the front end includes embedding the coordinate data into the LIDAR input. I am worried that separately infusing coordinate data may be adding unnecessary complexity to the NN architecture and even destructive in non line of sight (NLOS) cases? 
Aldebaro Klautau
10:04:55
Congratulations again to all participants of the PS-012 [beam-selection]! Great job!
Vishnu Ram
10:06:39
Of course, great job!
Mahdi Boloursaz Mashhadi
10:06:57
Thank you Aldebaro and your team for providing the Raymobtime dataset.
Mahdi Boloursaz Mashhadi
10:09:27
An lso for posing this Nice idea in the first place Aldebaro. Would be nice to collaborate to publish an overview kind of paper on this topic putting together achievements from all teams on the competition and your results of course.
Vishnu Ram
10:11:10
Tomorrow 15:30- 15:35 (CET), Call for papers: Ian Akyildiz, Editor-in-Chief, Special issue of ITU Journal on Future and Evolving Technologies (ITU J-FET)​): “AI/ML Solutions in 5G and Future Networks”
Vishnu Ram
10:12:10
17 Dec 15:30- 15:35 (CET), Call for papers: Ian Akyildiz, Editor-in-Chief, Special issue of ITU Journal on Future and Evolving Technologies (ITU J-FET)​): “AI/ML Solutions in 5G and Future Networks”
Ahmad Nagib
10:14:46
as far as I understand you are using some parameters to predict current throughput value,if so, did you consider predicting future throughput values and not the current ones based on historical data? (time series forecasting I mean)
Shyam Parekh
10:21:46
Are the presentations available?
Thomas Basikolo
10:22:46
The presentations are available on the website: https://www.itu.int/en/ITU-T/AI/challenge/2020/Pages/PROGRAMME.aspx
Loïck Bonniot
10:30:13
Nice talk! May I ask what kind of aggregations you used in graph convolutions?
Ahmad Nagib
10:30:21
what is the range of throughput values here?to get a sense of the RMSE!
Akihiro Nakao
10:33:22
I cannot speak right now, but what is the computational complexity of GCN? compared to CNN?
José Suárez-Varela
10:41:04
Regarding Loïck's question, I leave here an interesting paper that includes a comprehensive evaluation of different aggregation functions in GNN/GCN: https://arxiv.org/pdf/2004.05718.pdf
Thomas Basikolo (ITU)
10:42:49
We will continue tomorrow with presentations from various problem statements: https://www.itu.int/en/ITU-T/AI/challenge/2020/Pages/PROGRAMME.aspx
Thomas Basikolo (ITU)
10:43:02
For Participants in The ITU AI/ML in 5G Challenge, please respond to this survey:https://forms.office.com/Pages/ResponsePage.aspx?id=12TkI-YEh0uRPCS9iSGf0-yqkfLCoQ9IpTbc_XELf95URExJWlM2OExXVEQ5VFhFSTUzWUU3SDU1WS4uThe Grand Challenge Finale (Keynotes + Awards) will take place on 17 Dec.Please register here: https://itu.zoom.us/webinar/register/WN_bm4oJ_OeRH-Ilnb4GY_gMQ
mashael
10:48:05
R
Ahmad Nagib
10:48:22
will the recording be shared by any chance?
mashael
10:48:34
I vote to last team
Akihiro Nakao
10:48:39
thank you!
Vishnu Ram
10:48:49
Yes, recording will be available.
CARMELO JOSE ALBANEZ BASTOS FILHO
10:48:53
bye
Mengfei Feng
10:49:03
bye bye
Ahmad Nagib
10:49:14
OK great.. thanks Vishnu! :))
Raymond Francis Sarmiento
10:49:15
Looking forward to receiving the recording. Thank you
Cresting - Shiyi Zhu
10:49:16
bye