The research framework is organized around three complementary pillars: data-centric foundations addressing computational infrastructure and data management, model-centric foundations focused on algorithmic development and formal methods, and interdisciplinary applications spanning engineering, sciences, medicine, and humanities.

Cross-cutting methodological themes - uncertainty quantification, robustness, and reproducibility - are embedded throughout the framework to ensure research quality and conceptual coherence across all activities.

These areas form a mutually reinforcing ecosystem: advances in computational infrastructure enable sophisticated modelling approaches, methodological innovations expand application possibilities, and challenges from real-world deployments drive progress in both data strategies and algorithmic techniques. This integrated design supports research that is both theoretically rigorous and practically impactful.

1. Data-Centric AI and Data Science

Focus: Methods for acquiring, representing, managing, integrating, and processing data across heterogeneous sources and computational environments. This includes, but is not limited to, scalable data processing, distributed systems, data integration, data quality and validation, computational infrastructure, diagnostics, efficient and sustainable computation. 

2. Model-Centric AI and Data Science

Focus: Development, analysis, and evaluation of formal modelling frameworks for learning, inference, and prediction. This includes, but is not limited to, mathematical and algorithmic foundations, verification and validation, interpretability, explainability and human-AI interaction, statistical modelling, learning under uncertainty, robust and safe AI. 

3. Data Science and AI Applications

Focus: Application of AI and data science methods to problem-solving in engineering and scientific systems, medicine and life sciences, social sciences, and humanities.

Cross-Cutting Methodological Themes:

The following methodological priorities are addressed across all research areas: Uncertainty quantification, robustness, reproducibility, replicability, generalisability, and interpretability.