TinyML in Africa: Opportunities and Challenges (original) (raw)

2021 IEEE Globecom Workshops (GC Wkshps), 2021

Abstract

The integration of Internet of Things (IoT) and Machine Learning (ML) is driving the emergence of TinyML, a subset of ML for which models are constrained to be executed on low resource embedded devices. Currently, most IoT applications in Africa are based on cloud-centric architecture making it difficult to enjoy ML-driven intelligence due to connectivity, energy, and cost challenges making tinyML an ideal solution to such challenges. This paper presents an introduction to TinyML and discusses its general applications, solutions to concerns on the impact of Artificial Intelligence (AI) in the achievement of the sustainable development goals (SDGs), and the Information Technology for Development (ICT4D) TinyML technology requirements. TinyML applications classifications are highlighted with specific examples from Africa. Different challenges for the adoption of tinyML in Africa have also been presented with the opportunities that can arise from such challenges being shown. This paper shows how tinyML has great potential in Africa where the use of Embedded systems and AI is still underexploited.

Samson Ooko hasn't uploaded this paper.

Let Samson know you want this paper to be uploaded.

Ask for this paper to be uploaded.