Dr. Innocent Nyalala
Cultural Sustainability in the New Technological Age
Exploring the intersection of technology and culture, this research presents a framework for harmonising Al policies in East Africa while preserving cultural sustainability in a rapidly evolving technological age.
CLOVES-4603: Benchmarking Classical Texture Features and Fine-Tuned Deep Models for Clove Quality Grading.
We introduce CLOVES-4603, the first public benchmark for clove quality grading, comprising 4,603 images across four commercial grades. Fine-tuned deep models achieve up to 99.67% accuracy, with our classical texture baseline reaching 92.94%, providing deployment-ready trade-offs for agricultural settings with limited infrastructure.
Deep Learning for Automated Clove Quality Grading: A Feasibility Study Using CNN Architectures on a Novel Zanzibar Dataset
We investigate automated clove quality grading using deep learning on a novel dataset of 5,298 images collected at the Zanzibar State Trading Corporation. ResNet18 achieves 94.46% accuracy across four commercial grades, establishing a strong baseline for responsible AI deployment in East African agriculture.
Culturally Attuned and Resource-Aware Foundation Models for East African Agriculture: A Theoretical Framework and Research Agenda.
We propose CARA-FM, a theoretical framework for building foundation models suited to East African agriculture, where over 175 million people are underserved by AI systems designed without their constraints in mind. The framework addresses computational limitations, linguistic diversity, and indigenous knowledge systems across four pillars: community-driven data, edge-first model design, indigenous knowledge integration, and participatory governance.
Unifying Perspectives on Learning Biases: A Data-Centric Intervention for Holistic Fairness, Robustness, and Generalization
This paper explores data-centric approaches to simultaneously improve AI fairness, robustness, and generalization by treating learning biases as a unified challenge rather than isolated problems. It proposes a specialized intervention framework designed to build trustworthy systems that remain reliable and unbiased across different modalities.
AI Sovereignty in the Global South: Power, Dependency, and Strategic Futures
Analyzes structural power asymmetries in AI development across the Global South, proposing multipolar policy frameworks involving the African Union, ASEAN, and South-South cooperation to counter digital dependency on foreign platforms.
From Art to Algorithms: Co-Designing AI for Clove Grading with Zanzibar's Indigenous Experts
Proposes a socio-technical approach to modernizing Zanzibar's clove grading system. Validates a segmentation-first deep learning architecture, achieving 99.0% accuracy, while advocating for co-design with Indigenous experts to develop a low-cost mobile tool that fosters transparency and trust.
BAFNet: Deep contour-aware features for colorectal polyps segmentation
Introduces a boundary-aware feature fusion neural network for improved colorectal polyp segmentation through gating mechanisms that select high-quality contour features and enhance semantic accuracy.
Digital Sovereignty or Digital Serfdom? AI Strategies, Geopolitics, and the Quest for an Equitable Global South
This study examines whether Global South countries are advancing toward digital sovereignty or drifting into digital serfdom. It uses a seven-dimensional framework and four case studies (Singapore, Vietnam, Kenya, and Syria) to reveal asymmetries in computing, talent, and regulation.
Towards Culturally Attuned and Resource-Aware Foundation Models for East African Agriculture: A theoretical framework and research agenda
Presented theoretical framework for culturally attuned foundation models for East African agriculture.
How can Innovation Modernize the Public Service Delivery and Create Inclusive Digital Economy
A talk on innovation's role in modernizing public service delivery and fostering an inclusive digital economy.
Computing at the Edge of Possibility: Building Agricultural AI for Resource Constrained East Africa
Faculty talk on developing agricultural AI solutions for resource-constrained environments in East Africa.
Prime Directives for Responsible AI for Africa: A Manifesto for Inclusive Technology
We present a holistic framework for developing inclusive, ethical, and sustainable AI that addresses Africa's unique cultural, social, and economic contexts. We emphasize participatory design, data equity, and local empowerment to ensure AI benefits all African communities.
Rectifying the extremely weakened signals for cassava leaf disease detection
Developed an artificial intelligence approach to accurately detect cassava leaf diseases by enhancing weak signal patterns in agricultural imaging.
