Miguel-Ángel Fernández-Torres

Miguel-Ángel Fernández-Torres

Assistant Professor at Universidad Carlos III de Madrid in Leganés, Madrid, Spain , He/Him

ELLIS Member ITU-UN GI-AI4R WG-Data Co-Lead
About

Miguel-Ángel Fernández-Torres is Assistant Professor at Universidad Carlos III de Madrid (Dept. of Signal Theory and Communications) and member of the Audiovisual Intelligence Group (AIG). He is also a member of the European Laboratory for Learning and Intelligent Systems (ELLIS) and co-leads the Working Group on Data at the Global Initiative on Resilience to Natural Hazards through Artificial Intelligence Solutions (GI-AI4R) at the International Telecommunication Union (ITU) of the United Nations.

His current research within Artificial Intelligence for Earth and Climate Sciences focuses on the design and understanding of explainable deep generative models and attention mechanisms for anomaly and extreme event detection — including wildfires, droughts, and heatwaves. His broader interests span computer vision, remote sensing, machine and deep learning, image and video processing, explainable AI, and visual attention modeling.

He received the Audiovisual Systems Engineering degree, the M.S. in Multimedia and Communications, and the Ph.D. in Multimedia and Communications from Universidad Carlos III de Madrid (UC3M), Spain, in 2013, 2014, and 2019, respectively, and was a postdoctoral researcher at the Image and Signal Processing Group, Universitat de València, between 2020 and 2024. He has also been a Visiting Researcher at the Visual Perception Laboratory of Purdue University (2016), a Beyond Fellow at the International Future Lab AI4EO, Technische Universität München (2022), and a Visiting Researcher at the Department of Artificial Intelligence, Fraunhofer Institute for Telecommunications (HHI), Berlin (2025).

23 journal papers (17 Q1) · 14 conference papers · h-index 16 (Google Scholar) Download full CV (PDF) →
Publications
37 total
Nature Communications · 2026 Bridging the weather and climate divide with artificial intelligence Gustau Camps-Valls, Alberto Carrassi, Francisco de Melo Viríssimo, et al., Veronika Eyring, Miguel-Ángel Fernández-Torres (5/20)
Artificial Intelligence for the Earth Systems · 2026 On the Predictive Skill of Artificial Intelligence-based Weather Models for Extreme Events using Uncertainty Quantification Rodrigo Almeida, Noelia Otero, Miguel-Ángel Fernández-Torres (3/4), Jackie Ma
Earth System Dynamics · 2026 A multivariate analysis of atmospheric drivers for Western European heatwaves Aytaç Paçal, Birgit Hassler, Katja Weigel, Miguel-Ángel Fernández-Torres (4/6), Gustau Camps-Valls, Veronika Eyring
IEEE Geoscience and Remote Sensing Magazine · 2026 Converting Raw Data Into Actionable Information: A topical review of artificial intelligence, machine learning, and digital twins in disaster management Arif R. Albayrak, Monique M. Kuglitsch, Jürg Luterbacher, et al., Miguel-Ángel Fernández-Torres (11/15)
Water Resources Research · 2026 The maturation of AI in drought science: A review of trends, pitfalls, and priorities Guido Ascenso, Matteo Giuliani, Jorge Pérez-Aracil, et al., Andrea Castelletti, Miguel-Ángel Fernández-Torres (8/17)
Scientific Data · 2026 A large-scale, multitask, multisensory dataset for climate-aware crop monitoring in the US from 2018–2022 Adrian Höhl, Stella Ofori-Ampofo, Miguel-Ángel Fernández-Torres (3/5), Rıdvan Salih Kuz, Xiao Xiang Zhu
Show all 37 publications
2026
2025
2024
2022
Research Projects
  • 2026 - 2030
    CLOSER: Closing the LOop — Attribution of complex climate disaStERs to climatic and to policy changes HORIZON-RIA
    PI: Paolo Scussolini (VU Amsterdam) · PI at UC3M: Miguel-Ángel Fernández-Torres
  • 2025 - 2026
    SEED: Spatiotemporal Explainable Geospatial Artificial Intelligence for Extreme Events and Their Drivers UC3M Funding for Research Activities
    PI: Miguel-Ángel Fernández-Torres (UC3M) · PI
  • 2025 - 2028
    IDEALCV-CM: Improved Deep Learning for Computer Vision Regional Research Project, Comunidad de Madrid (TEC-2024/COM-322)
    PI: Marcelo Bertalmío (CSIC) · Researcher at UC3M
  • 2024 - 2026
    ThinkingEarth: Copernicus Foundation Models for a Thinking Earth HORIZON-EUSPA-2022-SPACE (101130544)
    PI: Ioannis Papoutsis (National Technical University of Athens) · Researcher at UVEG
  • 2020 - 2026
    USMILE: Understanding and Modelling the Earth System with Machine Learning European Research Council (ERC), grant 855187
    PI: Veronika Eyring (German Aerospace Center) · Researcher at UVEG
  • 2021 - 2025
    XAIDA: Extreme Events — Artificial Intelligence for Detection and Attribution HORIZON 2020 (101003469)
    PI: Robert Vautard (IPSL) and Dim Coumou (VU Amsterdam) · Researcher at UVEG
  • 2022 - 2024
    Deep Extremes: AI4Science Multi-Hazards, Compounds and Cascade Events European Space Agency (ESA)
    PI: Miguel Mahecha (Universität Leipzig) · Co-coordinator at UVEG
  • 2018 - 2020
    Saliencia y Atención: rePresentación, Interpretación y EmergeNcia National Research Project (TEC2017-84395-P)
    PI: Ascensión Gallardo Antolín and Carmen Peláez Moreno (UC3M) · Team Member
  • 2015 - 2017
    Saliencia y Atención: Multimodalidad, Context-Awareness, Auto-Adaptación y Bioinspiración National Research Project
    PI: Ascensión Gallardo Antolín and Carmen Peláez Moreno (UC3M) · Team Member
Work Experience
  • 2024 - Present
    Assistant Professor Universidad Carlos III de Madrid

    Dept. of Signal Theory and Communications, Escuela Politécnica Superior. Member of the Audiovisual Intelligence Group (AIG), ELLIS Unit Madrid, and co-lead of the Working Group on Data at the ITU-UN Global Initiative on Resilience to Natural Hazards through AI Solutions (GI-AI4R). Research on explainable deep generative models for anomaly and extreme event detection using Earth observation and climate data.

  • 2020 - 2024
    Postdoctoral Researcher Universitat de València

    Image and Signal Processing (ISP) Group. Worked on deep learning and generative models for drought monitoring, spatio-temporal gap filling of satellite reflectances, and multisensory Earth observation datasets, within EU projects USMILE (ERC), XAIDA (H2020), and the ESA Deep Extremes initiative.

  • 2018 - 2020
    Teaching Assistant Universidad Carlos III de Madrid

    Dept. of Signal Theory and Communications. Continued research on visual attention modeling and autonomous driving scenarios while teaching undergraduate courses.

  • 2014 - 2018
    Research Assistant (FPU Grant) Universidad Carlos III de Madrid

    PhD studies in Multimedia and Communications, funded by the Spanish Ministry of Education’s University Faculty Training (FPU) programme. Research on hierarchical probabilistic models for spatio-temporal visual attention. Visiting Researcher at the Visual Perception Laboratory, Purdue University, in 2016.

Speaking
  • 2025
    Session 3A: Artificial Intelligence for Disaster Risk Management EU Science for Preparedness Conference, Copernicus Emergency Management Service, Turin, Italy
  • 2025
    Artificial Intelligence for Modeling and Understanding Extreme Weather and Climate Events Seminar, Fraunhofer Institute for Telecommunications (HHI), Berlin, Germany
  • 2025
    Harnessing AI for Smarter Early Warning Systems Webinar, International Water Management Institute (IWMI), Online
  • 2025
    AI for Decision-Support + Policy LEAP Wallerstein Panel Series — AI + Extreme Weather Preparedness, Columbia University, New York, USA
  • 2025
    Advanced Artificial Intelligence for Extreme Event Analysis: Hands-on with the AIDE Toolbox Living Planet Symposium, European Space Agency (ESA), Vienna, Austria

    Hands-on training session.

  • 2024
    Recent Trends, Challenges and Limitations of Explainable AI in Remote Sensing XAI4CV Workshop, CVPR, Seattle, WA, USA

    Oral presentation, with Adrian Höhl, Ivica Obadic, Dario Oliveira, and Xiaoxiang Zhu.

  • 2023
    Extreme Event Monitoring, Everywhere, All at Once: Challenges and Strategies Seminar, Department of Astrophysics, University of Zurich, Zurich, Switzerland
  • 2016
    A Probabilistic Topic Approach for Context-Aware Visual Attention Modeling 14th International Workshop on Content-Based Multimedia Indexing (CBMI), Bucharest, Romania

    Oral presentation, with Iván González Díaz and Fernando Díaz De María.