Artificial Intelligence and Deep Learning

16/12/2024dev

Introduction

UMMISCO leverages AI to build frugal, explainable, and interpretable models addressing health, biodiversity, and environmental monitoring. Projects explore embeddings, self-supervised learning, symbolic AI, and LLMs tailored to Global South contexts.

Scientific Issues & Challenges

Interpretability

Designing trustworthy, transparent AI for scientists and decision-makers.

Data Challenges

Addressing low-resource, multimodal, or biased datasets (e.g., African languages, biosignals).

Architecture

Designing context-specific models for embedded use, frugal hardware, and multilingual processing.

Applications

Health

DeepECG4U, DeepIntegrOmics for early disease detection.

Biodiversity

MetaPlantCode, eCOL+, AIME for coral reefs and marine acoustics.

Language & Law

NAWRAS, ESPERANTO for African legal text and language processing.

Transport

AI-enhanced traffic monitoring in I-Maroc and simulation coupling with urban models.

Research Activities & Animations

– Weekly AI seminars (~30/year).

– Doctoral support and interdisciplinary mentoring.

– Coordinated projects with biodiversity, health, and language experts.

Research

Related Research

Read All Research

Theme 1

Agent-Based Mathematical and Computer Modeling

Our agent-based modeling simulates social, economic, and environmental scenarios, aiding policymakers in urban planning and public health by predicting outcomes and epidemic spread.

Theme 2

Artificial Intelligence and Deep Learning

AI and deep learning analyze large datasets to reveal patterns and support sustainable development, using satellite imagery for climate impact predictions and solutions in health and urban planning.

Theme 3

Sensors and Data Collection

Our sensor networks enable real-time monitoring of critical environmental metrics, such as air quality and water resources, throughout Southeast Asia. This data empowers researchers and communities to make better-informed decisions for a sustainable future.

Contact us

Get in touch with us

If you have any questions or need assistance, please fill out our contact form, and our team will respond promptly. We’re here to support you with any inquiries about our research, collaborations, or programs. We look forward to connecting with you!