Deep Learning on Real Geophysical Data: A Case Study for Distributed Acoustic Sensing Research
This work addresses the problem of efficiently processing real geophysical data for researchers in seismology, but it is incremental as it focuses on scaling and tuning existing methods.
The authors tackled the challenge of designing deep learning for real, large, and complex geophysical data by developing a finely-tuned and efficiently scaled classifier to identify usable energy from seismic data using Distributed Acoustic Sensing, achieving a training speed increase of over two orders of magnitude with 16 times more GPUs on a 50,000-image dataset.
Deep Learning approaches for real, large, and complex scientific data sets can be very challenging to design. In this work, we present a complete search for a finely-tuned and efficiently scaled deep learning classifier to identify usable energy from seismic data acquired using Distributed Acoustic Sensing (DAS). While using only a subset of labeled images during training, we were able to identify suitable models that can be accurately generalized to unknown signal patterns. We show that by using 16 times more GPUs, we can increase the training speed by more than two orders of magnitude on a 50,000-image data set.