@article{fdi:010092172,
  title = {{E}xploring deep learning for volcanic source inversion},
  author = {{U}roz, {L}. {L}. and {Y}an, {Y}. {J}. and {B}enoit, {A}. and {A}lbino, {F}. and {B}ouygues, {P}. and {G}iffard-{R}oisin, {S}ophie and {P}inel, {V}irginie},
  editor = {},
  language = {{ENG}},
  abstract = {{M}achine learning has demonstrated potentiality for challenging physical tasks, such as inverting complex mechanisms with important data limitations. {I}t is now competing with traditional methods that involve statistical and physical modeling. {T}hese methods face significant challenges, including long computation time, extensive prior knowledge requirements, and sensitivity to scarce and noisy data which limit their ability to generalize. {R}egarding these difficulties, this article aims to explore the potential deployment of a deep learning-based method to solve an inverse problem in volcanology, that is, to estimate the volume change and depth of a {M}ogi-type source model from surface displacement measurements. {S}imulated displacement samples are used to get rid of insufficient amounts of real data and a lack of ground truth. {P}articular efforts are devoted to proper data preparation, including proposing a semi-automatic technique for training, validation, and testing data sampling and investigating the impact of data distribution, data diversity, and noise. {R}eal data over the {S}uswa volcano are also used to further assess the performance of the proposed deep learning method. {R}esults with both synthetic and real data provide evidence to consider deep learning-based methods for geophysical inverse problems.},
  keywords = {{D}eformation ; {N}oise ; {T}raining ; {M}athematical models ; {D}ata models ; {D}eformable models ; {D}eep learning ; {M}agma ; {I}nverse problems ; {D}isplacement measurement inversion ; {M}ogi model ; volcanic modeling},
  booktitle = {},
  journal = {{IEEE} {T}ransactions on {G}eoscience and {R}emote {S}ensing},
  volume = {62},
  numero = {},
  pages = {4513809 [11 p.]},
  ISSN = {0196-2892},
  year = {2024},
  DOI = {10.1109/tgrs.2024.3494253},
  URL = {https://www.documentation.ird.fr/hor/fdi:010092172},
}