Dr. Jawad Fayaz

Dr. Jawad Fayaz

Senior Lecturer in Computer Science (Data Science & AI)

Director of AI-PRISM Lab

Associate Director of Centre for Environmental Intelligence

University of Exeter, UK

About

Dr. Fayaz is a Senior Lecturer in Computer Science (Data Science and Artificial Intelligence) at the University of Exeter, UK. His research lies at the intersection of machine learning, probabilistic modelling, and decision science, with a focus on spatiotemporal representation learning and predictive modelling in complex systems. He particularly develops computational frameworks for risk quantification, early warning, and sequential decision-making for extreme events—high-impact low-probability and high-probability long-horizon risks—in natural hazards, environmental systems, and critical infrastructure.

His expertise includes temporal and graph-based deep learning, representation learning, process-informed machine learning, and Bayesian inference, with a focus on designing scalable learning algorithms and interpretable models.

Currently, he is the Associate Director of the Centre for Environmental Intelligence and Programme Lead of the Google DeepMind Research Ready Programme at the University of Exeter, UK. He also serves as an Executive Board member of the UK Collaboratorium for Research on Infrastructure and Cities (UKCRIC) and is a Fellow of the Durham Institute of Research, Development, and Invention (DIRDI). In addition, he is a member of the editorial boards of three journals: i) International Journal of Disaster Risk Reduction, ii) Frontiers in Earth Science, and iii) Smart and Sustainable Built Environment.

AI-PRISM Lab

AI for Probabilistic Risk Intelligence and Spatiotemporal Modelling

Meet the lab
Graph-Based Learning Time-Series Analysis Process-Informed ML Reinforcement Learning Bayesian Inference Feature Engineering Early Warning Systems Probabilistic Hazard Analysis Health Monitoring Surrogate Modelling

Research Areas

Computational methods applied to risk, safety, and decision-making problems in complex physical systems.

Methods
  • Graph Neural Networks
  • Temporal Neural Networks
  • Physics-Informed Machine Learning
  • Representation Learning
  • Reinforcement Learning
Application Domains
  • Natural Hazards earthquakes, floods, heatwaves
  • Environmental Systems weather, agriculture, land-use
  • Infrastructural Systems water systems, wind turbines, buildings, bridges
Research overview — spatiotemporal learning and risk quantification

Education

Post-Doctoral Research

University College London, UK

Doctor of Philosophy (PhD)

University of California, Irvine, USA

Master of Science (MS)

University of California, Irvine, USA

Contact