Machine learning for geophysical applications

Abstract

In recent years, Deep Learning has drawn a lot of interest from the geophysical community. It can be used for a variety of geophysical employments, by providing simpler and easier solutions to problems. It addresses problems to do with excessive need of computational resources and time consumption from large amounts of data, as well as the curse of dimensionality. Over the years there has been great progress with Deep Learning methods. An advantage of using Deep Learning in certain applications is that it allows its users to use remote sensing, so they can collect and analyze data even from places that are difficult to be approached by humans. With the help of TensorFlow and Keras libraries, Deep Learning plays a part in the improvement of exploring surrounding environments.