Page 23 - Crowdsourcing AI and Machine Learning solutions for SDGs - ITU AI/ML Challenges 2024 Report

Crowdsourcing AI and Machine Learning solutions for SDGs

6 Winning solutions

Since its inception in 2020, the ITU AI/ML Challenge has awarded prizes to over 300 participants
across various competitions. In this section, we highlight some of the remarkable solutions
submitted to the ITU AI/ML Challenge.

Photo 1: Representative winners of 2023 ITU AI/ML Challenge presenting their experience during the 2024 AI for
Good workshops

6.1 AI/ML for 5G-Energy Consumption Modelling

The aim of this challenge was to reduce the energy consumption of 5G base stations using
AI/ML. The winning solution, which was featured and showcased at COP28 in Dubai, utilized
artificial neural networks combined with other techniques to achieve up to a 20% reduction
in energy consumption. This competition attracted over 1000 participants and received more
than 12,246 submissions from 83 countries, demonstrating the global interest and innovation
in optimizing energy use in 5G networks.

The Challenge was hosted on both the ITU platform (link here: https:// challenge .aiforgood .itu
.int/ match/ matchitem/ 83) and the Zindi platform (link here: https:// zindi .africa/ competitions/
aiml -for -5g -energy -consumption -modelling)

Figure 11: Winner announcement of AI/ML for 5G-Energy Consumption Modelling
challenge at COP28 in Dubai

15