Practice challenges
FINANCIAL INCLUSION IN AFRICA – IAIO PRACTICE CHALLENGE
CARBON DIOXIDE PREDICTION CHALLENGE – IAIO PRACTICE CHALLENGE
Please note: Both these challenges are password protected in order to provide a private learning environment solely for Olympiad participants. To access the challenges, please reach out to your team leader.
HERE ARE SOME ADDITIONAL QUICK VIDEOS THAT EXPLAIN CERTAIN ASPECTS OF THE ZINDI PLATFORM:
During the practical round of the IAIO, all laptops will be pre-loaded with Anaconda and a number of other popular IDEs, as well as Python and a number of libraries. We recommend that you familiarize yourself with Anaconda.
Anaconda Beginners Guide for Linux and Windows
COMPETITION FORMAT
Scientific Round
Date: 9th September 2024
Duration: 5 hours
Key details: This round will test participants’ knowledge and skills through a series of six questions. These questions will be similar to those covered during the preparatory webinars and will require the skills introduced in those sessions.
This round requires participants to demonstrate their understanding of AI concepts and their ability to apply these concepts to solve problems effectively. The Scientific Round is designed to reflect real-world scientific research, challenging participants to identify and address limitations in existing approaches.
Participants should be proficient in basic coding skills, familiar with common deep learning Python libraries, and have a solid understanding of machine learning fundamentals.
Preparation Round
Date: 10th September 2024
Duration: 2 hours
Key details: Participants will have the opportunity to test the examination environment on the provided laptops. Please refer to the detailed schedule for more information.
Practical Round
Date: 11th September 2024
Duration: 5 hours
Key Details: The practical round of the International AI Olympiad will be hosted on the Zindi platform (www.zindi.africa). This intermediate-level challenge is designed to test participants’ practical problem-solving and coding skills in a time-constrained environment.
Format: Individuals will work on laptops pre-configured with identical: coding environment, pre-loaded libraries, pre-loaded datasets, and (limited) internet connectivity
Environment: The competition will be run on the Zindi platform, which will provide a seamless and secure environment for the challenge. On the platform participants will have access to the description of the problem and a live leaderboard which will automatically evaluate the solutions. Participants will be able to make up to 20 solution submissions to the platform during the five hour period.
Programming Language: Python
Technical Specifications: Participants will be required to use a set of predefined common open source machine learning libraries. The list of libraries will be provided to participants two weeks in advance to ensure they are well-prepared. We recommend participants familiarize themselves with the Zindi platform, python, and the common machine learning libraries before the competition to maximize their performance. There will be a webinar and materials will be shared in advance to help participants prepare.
ORGANIZERS
