Reza M. Asiyabi

Reza M. Asiyabi

AI/ML for Earth Observation Scientist

Hi 👋, I’m Reza!
I build AI systems that try to understand the world rather than just fit the data.

I’m a researcher at the University of Edinburgh working on interpretable, robust, and physics-aware AI. My focus is on designing machine-learning models that combine theory, data, and real-world constraints, an approach shaped by years of tackling the unique challenges of Earth Observation.

A lot of my recent work has been in geospatial AI: from land-cover analysis to estimating biomass from radar, and from geospatial MLLMs complex-valued neural data compression. working with noisy, physics-coupled data forced me to build models that go beyond pattern-recognition and actually reason about the underlying system.

I’m currently a Postdoctoral Research Associate in the School of GeoSciences, funded by the UK National Centre for Earth Observation (NCEO) and UKRI SECO project. My research focuses on domain-aware deep learning to understand radar–vegetation interactions; Essentially teaching AI to interpret forests through the lens of backscatter physics.

I’ve published across leading ML/AI and EO venues, won competitive fellowships, and built methods that bridge theoretical depth with practical impact.

Before Edinburgh, I completed a Marie Curie–funded PhD at the University POLITEHNICA of Bucharest as part of the MENELAOS-NT project, developing complex-valued neural networks for SAR. I then joined Stanford University to work on the Market Activity Index (MAI), creating a remote sensing approach to detect local markets in low-income countries through high-resolution EO data.

I enjoy building systems that turn messy, real-world data into meaningful insight, whether in geospatial applications or broader AI problems, and I thrive in teams where different disciplines push each other forward.

👉 Explore my work.

News

  • [Dec 2025] Presenting my work on process-guided AI models for biomass estimation at the EurIPS 2025 conference - AICC workshop.
  • [Jun 2025] I will be presenting two presentations at the ESA Living Planet Symposium 2025, one on complex-valued neural data compression models for SAR raw data compression, and another one on domain-aware AI for biomass estimation.
  • [Sep 2024] I will be attending the UK National Earth Observation Conference 2024.
  • [Aug 2024] I joined the UK National Centre for Earth Observation (NCEO) and the University of Edinburgh as a postdoctoral research associate.
  • [Jan 2024] I joined the Market Activity Index (MAI) project at Stanford University, Center on Food Security and the Environment (FSI).
  • [Dec 2023] Doctoral Thesis Viva, Deep Learning for SAR Data in Presence of Adversarial Samples, National University of Science and Technology POLITEHNICA Bucharest PhD Viva.
  • [Sep 2023] International Conference on Content-based Multimedia Indexing (CBMI) 2023, Orleans, France CBMI 2023.
  • [Jul 2023] IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 2023, Pasadena, California, US IGARSS 2023.
  • [May 2023] IEEE GRSS High-Performance and Disruptive Computing in Remote Sensing (HDCRS) summer school, Reykjavík, Iceland HDCRS.
  • [Mar 2023] MenelaosNT Forum, Bucharest, Romania Forum.
  • [Jul 2022] European Conference on Synthetic Aperture Radar (EuSAR) 2022, Leipzig, Germany EuSAR 2022.
  • [Jul 2022] IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 2022, Kuala Lumpur, Malaysia IGARSS 2022.
  • [Jun 2022] Visiting Researcher @ Zentrum für Sensorsysteme (ZESS), University of Siegen, Germany ZESS.
  • [Feb 2022] PhD Forum Sensing Anywhere Innovation, University of Siegen, Germany Forum.
  • [Oct 2021] Visiting Researcher @ Zentrum für Sensorsysteme (ZESS), University of Siegen, Germany ZESS.

Contact me with any questions (email on the left).