eduzhai > Applied Sciences > Engineering >

Data-driven geophysics from dictionary learning to deep learning

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 73 pages

Abstract: Understanding the principles of geophysical phenomena is an essential andchallenging task. "Model-driven " approaches have supported the development ofgeophysics for a long time; however, such methods suffer from the curse ofdimensionality and may inaccurately model the subsurface. "Data-driven "techniques may overcome these issues with increasingly available geophysicaldata. In this article, we review the basic concepts of and recent advances indata-driven approaches from dictionary learning to deep learning in a varietyof geophysical scenarios. Explorational geophysics including data processing,inversion and interpretation will be mainly focused. Artificial intelligenceapplications on geoscience involving deep Earth, earthquake, water resource,atmospheric science, satellite remoe sensing and space sciences are alsoreviewed. We present a coding tutorial and a summary of tips for beginners andinterested geophysical readers to rapidly explore deep learning. Some promisingdirections are provided for future research involving deep learning ingeophysics, such as unsupervised learning, transfer learning, multimodal deeplearning, federated learning, uncertainty estimation, and activate learning.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×