Page 94 - ITU Journal Future and evolving technologies Volume 2 (2021), Issue 4 – AI and machine learning solutions in 5G and future networks

ITU Journal on Future and Evolving Technologies, Volume 2 (2021), Issue 4, 5 August 2021Technologies, Volume 2 (2021), Issue 4
ITU Journal on Future and Evolving

[26] F . Scarselli, M. Gori, A. C. Tsoi, M. Hagenbuchner and [38] Brownlee, J. (2016). “Deep learning with Python:
G. Monfardini, “The graph neural network model,” develop deep learning models on Theano and Tensor‑
IEEE Transactions on Neural Networks, pp. 61–80, Flow using Keras”. Machine Learning Mastery.
2008.
[39] Pedregosa, F ., et al. (2011). “Scikit‑learn: Machine
[27] P . Battaglia, J. Hamrick, V . Bapst, A. Sanchez‑ learning in Python”. The Journal of Machine Learning
Gonzalez, V . Zambaldi, M. Malinowski, A. Tacchetti, D. Research, 12, 2825‑2830.
Raposo, A. Santoro and R. Faulkner, “Relational induc‑
tive biases, deep learning, and graph networks”. arXiv [40] Yang, X., & Vaidya, N. (2005, March). “On physical
carrier sensing in wireless ad hoc networks”. In Pro‑
preprint arXiv:1806.01261, 2018.
ceedings IEEE 24th Annual Joint Conference of the
[28] M. Fey and J. E. Lenssen, “Fast Graph Representation IEEE Computer and Communications Societies.
Learning with PyTorch Geometric,” in ICLR Workshop (Vol. 4, pp. 2525‑2535). IEEE.
on Representation Learning on Graphs and Mani‑ [41] Wilhelmi, F ., Barrachina‑Muñoz, S., Bellalta, B., Cano,
folds, 2019.
C., Jonsson, A., & Ram, V . (2020). “A lexible machine‑
[29] P . Soto, D. Góez, M. Camelo, N. Gaviria and K. Mets, learning‑aware architecture for future WLANs”. IEEE
“ITU AI/ML in 5G Challenge 2020: Team ATARI Communications Magazine, 58(3), 25‑31.
(Github repository)”, 12 Nov 2020. Online: https://
[42] Wilhelmi, F ., Carrascosa, M., Cano, C., Jonsson, A.,
github.com/ITU‑AI‑ML‑in‑5G‑Challenge/
Ram, V ., & Bellalta, B. (2020). “Usage of Network
ITU‑ML5G‑PS‑013‑ATARI. [Accessed 20 Feb 2021]. Simulators in Machine‑Learning‑Assisted 5G/6G Net‑
works”. IEEE Wireless Communications Magazine,
[30] Murtagh, F . (1991). “Multilayer perceptrons for clas‑
si ication and regression”. Neurocomputing, 2(5‑6), 2021.
183‑197.
[31] R. Vallés (2020), “MLP Throughput prediction AUTHORS
(Github repository)”. Online: https:// Francesc Wilhelmi holds a Ph.D.
github.com/VPRamon/MLP‑Throughput‑prediction. in information and communication
[Accessed 20 Feb 2021].
technologies (2020) from Universi‑
[32] Chen, T ., & Guestrin, C. (2016, August). “Xgboost: A tat Pompeu Fabra (UPF). He is cur‑
scalable tree boosting system”. In Proceedings of the rently a postdoctoral researcher in
the Mobile Networks department at
22nd acm sigkdd international conference on knowl‑
edge discovery and data mining (pp. 785‑794). Centre Tecnològic de Telecomunicacions de Catalunya
(CTTC) and a teaching assistant at UPF and at Universi‑
[33] Yeo, I‑K, and R Johnson. 2000. “A New Family tat Oberta de Catalunya (UOC).
of Power Transformations to Improve Normality or
Symmetry.” Biometrika 87 (4): 954–59. David Góez is a telecommuni‑
cations engineer from the ITM
[34] Kuhn M, Wickham H (2020). “Tidymodels: a collec‑ Metropolitan Technological In‑
tion of packages for modeling and machine learning stitute, a Master in Automation
using tidyverse principles”. Available and Industrial Control from the
online: https: //www.tidymodels.org. ITM, a PhD student in electronic
and computer engineering at the
[35] M Alfai i, A Algunayah, A Aloshan, M Abid & K
Sahari (2020). “ITU 2020: Problem 13 ‑ STC Team 2”. University of Antioquia. His research focuses on radio‑
Github repository. Available online: https://github. communication systems whose signal processing blocks
com/ITU‑AI‑ML‑in‑5G‑Challenge/p13_stc_team2. can be built with machine learning.
[Accessed on 21 February 2021] Paola Soto is currently pursuing
her Ph.D. degree at University of
[36] Jorge Martin Pérez & Luigi Girletti (2020).
Antwerp ‑ imec, Belgium. Her doc‑
“ITU‑ML5G‑PS‑013”, Github repository. Avail‑
able online: https://github.com/MartinPJorge/ toral research investigates machine
ITU‑ML5G‑PS‑013. [Accessed on 21 February 2021] learning techniques applied to net‑
work management.
[37] Rajasekar Mohan, K Venkat Ramnan, Megha G She received a B.Sc. degree in Electronics and an M.Sc. de‑
Kulkarni, and Vishalsagar Udupi (2020), “Net Intels gree in telecommunications engineering from the Univer‑
solution for the ITU‑T AI Challenge (Github reposi‑ sity of Antioquia, Medellı́n, Colombia, in 2014 and 2018,
tory)”. Online: https://github.com/venkatramnank/ respectively. Her main research interests include arti i‑
UPF_ITU_T_5G_ML_AI_Challenge_NetIntels. cialintelligence, machinelearning, networkmanagement,
[Accessed 20 Feb 2021]. and resource allocation algorithms.

78 © International Telecommunication Union, 2021