About
I am a researcher at Orion Lab, a joint research group of the National Observatory of Athens and the National Technical University of Athens. My work focuses on the application of deep learning techniques to Earth Observation. I am currently nearing the completion of my PhD in Uncertainty-Aware Deep Learning for Earth Observation at the University of Valencia under the supervision of Gustau Camps-Valls and Ioannis Papoutsis. I am also co-founder of Manteo AI, a chat-map assistant that simplifies access to Earth Observation data and insights.
Research
My research lies at the intersection of Computer Vision, Uncertainty Quantification, and Earth Observation. I’m particularly motivated by the potential of AI to help address pressing environmental challenges, especially in areas like disaster management, natural hazards, and climate change. During my PhD, I worked on developing deep learning models for wildfire danger forecasting and explored uncertainty estimation in Earth Observation deep learning models to improve their reliability and trustworthiness. I am also interested in broader Computer Vision topics, particularly representation learning and multi-modal models, and their application in Earth Observation for extracting deeper insights from satellite data.
Background
I hold a BSc in Mathematics from the National and Kapodistrian University of Athens (2013), followed by an MSc in Applied Mathematics from the same institution (2016). I then completed a second MSc, in Data Science, from Athens University of Economics and Business in 2017. From 2018 to 2020, I worked as a Data Scientist at Omilia Conversational Intelligence, a company developing conversational AI systems, where I led the Data Science Team. My main research projects there involved text data augmentation combining deep learning with rule-based methods, text clustering and classification with LSTMs and Transformers, and text representation learning with variational autoencoders.
News
- July 2025: I attended Greeks in AI Symposium, where we had the opportunity to present Manteo AI and interact with the Greek community in AI.
- June 2025: Our paper On the Generalization of Representation Uncertainty in Earth Observation was accepted at ICCV 2025.
- May 2025: My colleague Nikolaos Ioannis Bountos presented our work “On the Generalization Capacity of Pretrained Uncertainties in Remote Sensing” at the ESA-NASA International Workshop on AI Foundation Models for EO
- April 2025: Our paper Probabilistic Machine Learning for Noisy Labels in Earth Observation is uploaded to arXiv.
- February 2025: Our paper Artificial intelligence for modeling and understanding extreme weather and climate events is published at Nature Communications.