Page 21 - Crowdsourcing AI and Machine Learning solutions for SDGs - ITU AI/ML Challenges 2024 Report

Crowdsourcing AI and Machine Learning solutions for SDGs

5 Problem statements

Participants of the ITU AI/ML Challenge can solve real-world problems (including those
with social relevance). Problem statements are contributed either from ITU’s standards and
specifications, or from hosts of problem statements who are institutions interested in advancing
SDGs or can be decided by the participant(s) themselves. Problem statements will fall into a
specific challenge domain based on the problem owner (host) interest and resources.

The AI for Good Global Summit identifies practical applications of AI/ML with the potential to
accelerate progress towards the United Nations Sustainable Development Goals. Solutions are
invited in fields such as education, healthcare and wellbeing, social and economic equality,
climate action, natural disaster management, space, and smart and safe mobility. Selected
teams will be invited to participate in the AI for Good Summit.

Figure 10: Sample Challenge problem statements

The ITU AI/ML Challenge continues to host problem statements from hosts around the world.
Some of the scheduled problem statements are as follows:

• Green Telecom: Smart Energy Supply Scheduling [Smart energy supply scheduling for
both carbon footprint reduction and network reliability guarantee]
• Beam-level Traffic Prediction
• Specializing Large Language Models for Telecom Networks
• Ground-level NO2 Estimation Challenge
• Radio Resource Management (RRM) for 6G in-X Subnetworks
The ITU AI/ML Challenge serves as a crucial bridge between current innovations and future
research and standards. By engaging participants in solving real-world problems using AI
and ML, the challenge fosters the development of practical solutions that can inform future
research directions. These solutions often lead to new insights and discoveries, fuelling further
investigations and academic studies.

13