Skip to main content
Log in

Deep Artificial Neural Networks as a Tool for the Analysis of Seismic Data

  • Published:
Seismic Instruments Aims and scope Submit manuscript

Abstract

The number of seismological studies based on artificial neural networks has been increasing. However, neural networks with one hidden layer have almost reached the limit of their capabilities. In the last few years, there has been a new boom in neuroinformatics associated with the development of third-generation networks, deep neural networks. These networks operate with data at a higher level. Unlabeled data can be used to pretrain the network, i.e., there is no need for an expert to determine in advance the phenomenon to which these data correspond. Final training requires a small amount of labeled data. Deep networks have a higher level of abstraction and produce fewer errors. The same network can be used to solve several tasks at the same time, or it is easy to retrain it from one task to another. The paper discusses the possibility of applying deep networks in seismology. We have described what deep networks are, their advantages, how they are trained, how to adapt them to the features of seismic data, and what prospects are opening up in connection with their use.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  • Böse, M., Wenzel, F., and Erdik, M., PreSEIS: A Neural Network-Based Approach to Earthquake Early Warning for Finite Faults, Bull. Seismol. Soc. Am., 2008, vol. 98, no. 1, pp. 366–382.

    Article  Google Scholar 

  • Cybenko, G., Approximation by superpositions of a sigmoidal function, Math. Control Signals Syst., 1989, no. 2, pp. 303–314.

    Article  Google Scholar 

  • Friedman, J.H., An overview of predictive learning and function approximation, in From Statistics to Neural Networks, vol. 136 of NATO ASI Series, Cherkassky, V., Friedman, J.H., and Wechsler, H., Eds., Berlin: Springer, 1994, pp. 1–61.

    Google Scholar 

  • Gravirov, V.V. and Kislov, K.V., The seismic data preparation program for training of an artificial neural network for ultrashort earthquake warning system, in Book of Abstracts, 10th International Conference “Problems of Geocosmos”, St. Petersburg, 2014, pp. 86–87. http://geo.phys.spbu.ru/ geocosmos/book_of_abstracts.pdf. Accessed December 31, 2014.

    Google Scholar 

  • Gravirov, V.V. and Kislov, K.V., DataCollector: Software for Compiling the Output Data Volumes to Learn Neural Networks in the Matlab System, State Software Registration Certificate no. 2015611797, 2015.

    Google Scholar 

  • Gravirov, V.V., Kislov, K.V., and Vinberg, F.E., The informative signal separation from the high noise level non-stationary seismic data with the use of the neural network classifiers, Ind. Autom. Control Syst. Controllers, 2012, no. 12, pp. 55–59.

    Google Scholar 

  • Gravirov, V.V., Kislov, K.V., Gravirova, L., and Vinberg, F., The use of wavelet transformation techniques in structure of an artificial neural network for recognition of early arrival of earthquakes on strongly noisy seismic records, in Book of Abstracts CTBT: Science and Technology 2013, Vienna, Austria, 2013, p. 139. http://www.ctbto.org/fileadmin /user_upload/ SnT2013/bookofabstracts.pdf. Accessed December 31, 2013.

    Google Scholar 

  • Hinton, G.E., Learning multiple layers of representation, Trends Cognit. Sci., 2007, no. 10, pp. 428–434.

    Article  Google Scholar 

  • Hinton, G.E., Osindero, S., and The, Y.-W., A fast learning algorithm for deep belief nets, Neural Comput., 2006, vol. 18, pp. 1527–1554.

    Article  Google Scholar 

  • Hochreiter, S., Bengio, Y., Frasconi, P., and Schmidhuber, J., Gradient flow in recurrent nets: The difficulty of learning long-term dependencies, in A Field Guide to Dynamical Recurrent Networks, Kolen, J.F. and Kremer, S., Eds., Los Alamitos, Calif.: IEEE Press, 2001.

    Google Scholar 

  • Holdaway, K.R., Advanced seismic attribute analysis: Deep learning, Finding Petroleum forum “Transforming Subsurface Interpretation”, London, 2015. http://80f54e3872b7907eeda8- 4de6a2a3e09efd2caa4972e2c52e8eac.r36.cf1.rackcdn. com/keith%20holdaway.pdf. Accessed December 31, 2015.

    Google Scholar 

  • Kislov, K.V. and Gravirov, V.V., Use of artificial neural networks for classification of noisy seismic signals, Seism. Instrum., 2017, vol. 53, no. 1, pp. 87–101.

    Article  Google Scholar 

  • Köhler, A., Ohrnberger, M., and Scherbaum, F., Unsupervised pattern recognition in continuous seismic wavefield records using Self-Organizing Maps, Geophys. J. Int., 2010, vol. 182, no. 3, pp. 1619–1630.

    Article  Google Scholar 

  • Längkvist, M., Karlsson, L., and Loutfi, A., A review of unsupervised feature learning and deep learning for timeseries modeling, Pattern Recognit. Lett., 2014, vol. 42, no. 1, pp. 11–24.

    Article  Google Scholar 

  • Le Roux, N. and Bengio, Y., Representational power of restricted Boltzmann machines and deep belief networks, Neural Comput., 2008, vol. 20, pp. 1631–1649.

    Article  Google Scholar 

  • Lin, Chu-Chieh J., Lin, Pei-Yang, Chang, Tao-Ming, Lin, Tzu-Kun, Weng, Yuan-Tao, Chang, Kuo-Chen, and Tsai, Keh-Chyuan, Development of on-site earthquake early warning system for Taiwan, in Earthquake Research and Analysis–New Frontiers in Seismology, D’Amico, S., Ed., InTech, 2012, Ch. 13, pp. 329–358.

    Google Scholar 

  • Madureira, G. and Ruano, A.E., A neural network seismic detector, Acta Tech. Jaurinensis, 2009, vol. 2, no. 2, pp. 159–170.

    Google Scholar 

  • Mikolov, T., Karafiat, M., Burget, L., Cernocky, J., and Khudanpur, S., Recurrent neural network based language model, 11th Annual Conference of the International Speech Communication Association, Makuhari, Chiba, Japan, 2010, pp. 1045–1048.

    Google Scholar 

  • Perervenko, V., The third generation of neural networks: Deep networks, 2014. https://www.mql5.com/ru/articles/1103#1_4. Accessed December 31, 2014.

    Google Scholar 

  • Schmidhuber, J., Deep learning in neural networks: An overview, Neural Networks, 2015, vol. 61, pp. 85–117.

    Article  Google Scholar 

  • Tat’yankin, V.M. and Dyubko, I.S., Deep confidence neural networks compared to multilayer perceptron, Vestn. Yugorsk. Gos. Univ., 2015, no. 2, pp. 87–89.

    Google Scholar 

  • Wiskott, L. and Sejnowski, T.J., Slow feature analysis: Unsupervised learning of invariances, Neural Comput., 2002, vol. 14, pp. 715–770.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to K. V. Kislov.

Additional information

Original Russian Text © K.V. Kislov, V.V. Gravirov, 2017, published in Seismicheskie Pribory, 2017, Vol. 53, No. 1, pp. 17–28.

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kislov, K.V., Gravirov, V.V. Deep Artificial Neural Networks as a Tool for the Analysis of Seismic Data. Seism. Instr. 54, 8–16 (2018). https://doi.org/10.3103/S0747923918010073

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.3103/S0747923918010073

Keywords

Navigation