This project will develop advanced machine learning models for prediction, modeling, emulation, detection and causal inference. Three main application and development domains will be:
- Physics-aware and explainable machine learning for dust and cloud properties statistical retrieval of dust aerosols and cloud properties.;
- Automatic detection of aerosol-cloud interactions in observations space by designing statistical emulators of high-resolution simulations, which will allow automatic sensitivity analysis and fast retrievals;
- Detection of aerosol-cirrus cloud interactions in satellite data via unsupervised learning and observational causal inference to identify latent factors potentially impacting the detection.
Gustau Camps-Valls (supervisor), Philip Stier (co-supervisor), Ralf Giering (non-academic advisor)
6 months at the University of Oxford and the Alan Turing Institute & 3 months at FastOpt
Enrolment in Doctoral degree
PhD in Image Processing at Universitat de Valencia