Skip to main content

Curriculum Learning in Deep Neural Networks for Financial Forecasting

  • Conference paper
  • First Online:
Mining Data for Financial Applications (MIDAS 2019)

Part of the book series: Lecture Notes in Computer Science ((LNAI,volume 11985))

Included in the following conference series:

Abstract

For any financial organization, computing accurate quarterly forecasts for various products is one of the most critical operations. As the granularity at which forecasts are needed increases, traditional statistical time series models may not scale well. We apply deep neural networks in the forecasting domain by experimenting with techniques from Natural Language Processing (Encoder-Decoder LSTMs) and Computer Vision (Dilated CNNs), as well as incorporating transfer learning. A novel contribution of this paper is the application of curriculum learning to neural network models built for time series forecasting. We illustrate the performance of our models using Microsoft’s revenue data corresponding to Enterprise, and Small, Medium & Corporate products, spanning approximately 60 regions across the globe for 8 different business segments, and totaling in the order of tens of billions of USD. We compare our models’ performance to the ensemble model (of traditional statistics and machine learning) currently being used by Microsoft Finance. Using this in-production model as a baseline, our experiments yield an approximately 30% improvement overall in accuracy on test data. We find that our curriculum learning LSTM-based model performs best, which shows that one can implement our proposed methods without overfitting on medium-sized data.

Supported by Microsoft Corp., where all research was conducted.

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

Access this chapter

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 39.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 54.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

References

  1. Barker, J., Gajewar, A., Golyaev, K., Bansal, G., Conners, M.: Secure and automated enterprise revenue forecasting. In: The Thirtieth AAAI Conference on Innovative Applications of Artificial Intelligence, pp. 7657–7664 (2018)

    Google Scholar 

  2. Bengio, Y., Louradour, J., Collobert, R., Weston, J.: Curriculum learning. In: Proceedings of the 26th Annual International Conference on Machine Learning, pp. 41–48. ACM (2009)

    Google Scholar 

  3. Borovykh, A., Bohte, S., Oosterlee, C.: Conditional time series forecasting with convolutional neural networks (2018)

    Google Scholar 

  4. Cirik, V., Hovy, E., Morency, L.P.: Visualizing and understanding curriculum learning for long short-term memory networks (2016)

    Google Scholar 

  5. Cleveland, R., Cleveland, W., McRae, J., Terpenning, I.: STL: a seasonal-trend decomposition procedure based on loess. J. Off. Stat. 6(1), 3–73 (1990)

    Google Scholar 

  6. Gajewar, A.: Improving regional revenue forecasts using product hierarchy. In: 39th International Symposium on Forecasting (2018)

    Google Scholar 

  7. Graves, A., Bellemare, M.G., Menick, J., Munos, R., Kavukcuoglu, K.: Automated curriculum learning for neural networks (2017)

    Google Scholar 

  8. Fawaz, H.I., Forestier, G., Weber, J., Idoumghar, L., Muller, P.A.: Transfer learning for time series classification (2018)

    Google Scholar 

  9. Flunkert, V., Salinas, D., Gasthaus, J.: DeepAR: probabilistic forecasting with autoregressive recurrent networks CoRR abs/1704.04110, (2017)

    Google Scholar 

  10. Gupta, P., Malhotra, P., Vig, L., Shroff, G.: Transfer learning for clinical time series analysis using recurrent neural networks. In: Medicine and Healthcare Workshop at ACM KDD 2018 Conference (2018)

    Google Scholar 

  11. Laptev, N., Yosinski, J., Li, L.E., Smyl, S.: Time-series extreme event forecasting with neural networks at uber. In: ICML 2017 Time Series Workshop, Sydney, Australia (2017)

    Google Scholar 

  12. Sirignano, J., Cont, R.: Universal features of price formation in financial markets: perspectives from deep learning (2018)

    Google Scholar 

  13. Spitkovsky, V.I., Alshawi, H., Jurafsky, D.: From babysteps to leapfrog: how less is more in unsupervised dependency parsing. In: Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, pp. 751–759. Association for Computational Linguistics (2010)

    Google Scholar 

  14. Sutskever, I., Vinyals, O., Le, Q.V.: Sequence to sequence learning with neural networks. In: Advances in Neural Information Processing Systems, vol. 27, pp. 3104–3112. Curran Associates, Inc. (2014)

    Google Scholar 

  15. Talagala, A., Hyndman, R., Athanasopoulos, G.: Meta-learning how to forecast time series. Monash Econometrics and Business Statistics Working Papers (2018)

    Google Scholar 

  16. Time Series Prediction with LSTM Recurrent Neural Networks in Python with Keras. https://machinelearningmastery.com/time-series-prediction-lstm-recurrent-neural-networks-python-keras. Accessed 19 Mar 2019

  17. TimeSeries Seq2Seq. https://github.com/JEddy92/TimeSeries_Seq2Seq. Accessed 19 Mar 2019

  18. van den OordSander, A., et al.: WaveNet: a generative model for raw audio (2016)

    Google Scholar 

Download references

Acknowledgements

We thank Kimyen Nguyen for her generous help with running experiments on security compliant machines considering the sensitivity of the finance data. We also thank Barbara Stortz, Deependra Hamal, and Mindy Yamamoto for their support of this project. The work of A.K. is jointly supported by Microsoft and the National Science Foundation Graduate Research Fellowship under Grant No. DGE – 1656518. Any opinion, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Allison Koenecke .

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 2020 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Koenecke, A., Gajewar, A. (2020). Curriculum Learning in Deep Neural Networks for Financial Forecasting. In: Bitetta, V., Bordino, I., Ferretti, A., Gullo, F., Pascolutti, S., Ponti, G. (eds) Mining Data for Financial Applications. MIDAS 2019. Lecture Notes in Computer Science(), vol 11985. Springer, Cham. https://doi.org/10.1007/978-3-030-37720-5_2

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-37720-5_2

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-37719-9

  • Online ISBN: 978-3-030-37720-5

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics