Scientists in Africa are pioneering innovative solutions to bridge the data gap, enabling AI models to function offline on low-end devices. The latest breakthrough involves training an AI model to serve both health and agriculture sectors with limited resources.
Meet **Ayobami Oyeleke**, a master’s student in Artificial Intelligence at the Ladoke Akintola University of Technology in Nigeria. Ayobami’s ambitious project aimed to create a dual-domain AI model capable of operating offline on devices with as little as 2GB RAM. This feat is crucial for addressing the significant data gap in Africa, where many healthcare facilities and rural areas struggle with limited connectivity.
Ayobami’s AI model combines two pre-trained architectures: RandomForest for health and MobileNetV3 for agriculture.
Ayobami’s solution leverages two pre-trained models: **RandomForest** for health applications and **MobileNetV3** for agriculture. By combining these, he created a single model capable of handling both domains. This innovative approach allows the model to provide valuable insights in healthcare and agriculture without relying on internet connectivity.
The potential impact of this breakthrough extends beyond Africa, as similar challenges exist in other developing regions.
What this means: The success of Ayobami’s project highlights the potential for offline AI solutions to improve the lives of people in data-constrained environments. By enabling AI models to function on low-end devices, researchers can create life-saving applications for healthcare and agriculture, ultimately contributing to sustainable development in Africa and beyond.
