Skip to main content
Springer Nature Link
Log in

A Fast and Robust Photometric Redshift Forecasting Method Using Lipschitz Adaptive Learning Rate

  • Conference paper
  • First Online:
Neural Information Processing (ICONIP 2022)

Part of the book series: Communications in Computer and Information Science ((CCIS,volume 1792))

Included in the following conference series:

  • 867 Accesses

Abstract

With the recent large astronomical survey experiments using high-resolution cameras and telescopes, there has been a tsunami of astronomical data that has been collected and is being utilized for important analysis. Based on pure photometric information, Redshift estimation is a crucial task of cosmology. The application of neural networks (NN) in this area is gaining popularity of late as NN performs well for large training samples. In this paper, we use Mean Absolute Error (MAE), as a metric, with a neural network to estimate the redshift of galaxies and quasars and show that MAE can be used as an alternate metric for this regression task. This paper uses Lipschitz constant based adaptive learning rate that involves hessian-free computation for faster training of the neural network. Results show that an adaptive learning rate based neural network with MAE converges much faster compared to a constant learning rate and reduces training time while providing MAE of 0.28 and Normalized Median Absolute Deviation (NMAD) is 0.03 for a data sample of 5 lakhs.

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

Access this chapter

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

Notes

  1. 1.

    Note: Since MAE does not support twice differentiable nature, but the functions discussed in Sect. 3.1 are twice differentiable, \(h\in D^2\), thus supports the assumption of the proof discussed in Sect. 3.

References

  1. Amendola, L., et al.: Cosmology and fundamental physics with the Euclid satellite. Living Rev. Relativ. 21, 1–345 (2018)

    Article  Google Scholar 

  2. Abbott, T., et al.: The dark energy survey: more than dark energy-an overview. Mon. Not. Royal Astron. Soc. 460(2), 1270–1299 (2016)

    Article  Google Scholar 

  3. de Jong, J.T.A., et al.: The kilo-degree survey. Exp. Astron. 35, 25–44 (2013). https://doi.org/10.1007/s10686-012-9306-1

    Article  Google Scholar 

  4. Chai, T., Draxler, R.R.: Root mean square error (RMSE) or mean absolute error (MAE)?–arguments against avoiding RMSE in the literature. Geosci. Model Dev. 7(3), 1247–1250 (2014)

    Article  Google Scholar 

  5. Saha, S., Prashanth, T., Aralihalli, S., Basarkod, S., Sudarshan, T.S.B., Dhavala, S.S.: LALR: theoretical and experimental validation of Lipschitz adaptive learning rate in regression and neural networks. arXiv preprint arXiv:2006.13307 (2020)

  6. Reza, M., Haque, M.A.: Photometric redshift estimation using ExtraTreesRegressor: galaxies and quasars from low to very high redshifts. Astrophys. Space Sci. 365(3), 1–9 (2020). https://doi.org/10.1007/s10509-020-03758-w

    Article  Google Scholar 

  7. Dalarsson, M., Dalarsson, N.: Tensors, Relativity, and Cosmology. Academic Press, Cambridge (2015)

    MATH  Google Scholar 

  8. Sen, S., Agarwal, S., Chakraborty, P., Singh, K.P.: Astronomical big data processing using machine learning: a comprehensive review. Exp. Astron. 53(1), 1–43 (2022). https://doi.org/10.1007/s10686-021-09827-4

    Article  Google Scholar 

  9. Sen, S., Saha, S., Chakraborty, P., Singh, K.P.: Implementation of neural network regression model for faster redshift analysis on cloud-based spark platform. In: Fujita, H., Selamat, A., Lin, J.C.-W., Ali, M. (eds.) IEA/AIE 2021. LNCS (LNAI), vol. 12799, pp. 591–602. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-79463-7_50

    Chapter  Google Scholar 

  10. Sandeep, V.Y., Sen, S., Santosh, K.: Analyzing and processing of astronomical images using deep learning techniques. In: IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT) (2021)

    Google Scholar 

  11. Monisha, et al.: An approach toward design and implementation of distributed framework for astronomical big data processing. In: Udgata, S.K., Sethi, S., Gao, X.Z. (eds.) Intelligent Systems. Lecture Notes in Networks and Systems, vol. 431, pp. 267–275. Springer, Singapore (2022). https://doi.org/10.1007/978-981-19-0901-6_26

    Chapter  Google Scholar 

  12. Mayank, K., Sen, S., Chakraborty, P.: Implementation of cascade learning using apache spark. In: 2022 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT). IEEE (2022)

    Google Scholar 

  13. Borne, K.D.: Astroinformatics: a 21st century approach to astronomy. arXiv preprint arXiv:0909.3892 (2009)

  14. Connolly, A.J., et al.: Slicing through multicolor space: galaxy redshifts from broadband photometry. arXiv preprint astro-ph/9508100 (1995). connolly1995slicing

    Google Scholar 

  15. Viquar, M., et al.: Emerging technologies in data mining and information security, machine learning in astronomy: a case study in quasar-star classification. In: Abraham, A., Dutta, P., Mandal, J., Bhattacharya, A., Dutta, S. (eds.) Emerging Technologies in Data Mining and Information Security. Advances in Intelligent Systems and Computing, vol. 814, pp. 827–836. Springer, Singapore (2019). https://doi.org/10.1007/978-981-13-1501-5_72

    Chapter  Google Scholar 

  16. https://keras.io/api/optimizers/

  17. Sarkar, J., Bhatia, K., Saha, S., Safonova, M., Sarkar, S.: Mon. Not. Royal Astron. Soc. 510 (2022)

    Google Scholar 

  18. Wilson, D., et al.: Photometric redshift estimation with galaxy morphology using self-organizing maps. Astrophys. J. 888, 33 (2020)

    Article  Google Scholar 

  19. Takase, T., et al.: Effective neural network training with adaptive learning rate based on training loss. Neural Netw. 101, 68–78 (2018)

    Article  Google Scholar 

  20. Xu, Z., Dai, A.M., Kemp, J., Metz, L.: Learning an adaptive learning rate schedule. arXiv preprint arXiv:1909.09712 (2019)

  21. Park, J., Yi, D., Ji, S.: A novel learning rate schedule in optimization for neural networks and it’s convergence. Symmetry 12, 660 (2020)

    Article  Google Scholar 

  22. Mediratta, I., Saha, S., Mathur, S.: LipARELU: ARELU networks aided by Lipschitz acceleration. In: 2021 International Joint Conference on Neural Networks (IJCNN). IEEE (2021)

    Google Scholar 

  23. Reddi, S.J., Kale, S., Kumar, S.: arXiv preprint arXiv:1904.09237 (2019)

  24. Luo, L., Xiong, Y., Liu, Y., Sun, X.: Adaptive gradient methods with dynamic bound of learning rate. arXiv preprint arXiv:1902.09843 (2019)

  25. Zhou, Z., et al.: AdaShift: decorrelation and convergence of adaptive learning rate methods. arXiv preprint arXiv:1810.00143 (2018)

  26. Smith, L.N., Topin, N.: Super-convergence: very fast training of neural networks using large learning rates. International Society for Optics and Photonics (2019)

    Google Scholar 

  27. Smith, L.N.: A disciplined approach to neural network hyper-parameters: part 1-learning rate, batch size, momentum, and weight decay. arXiv preprint arXiv:1803.09820 (2018)

  28. Liu, L., et al.: On the variance of the adaptive learning rate and beyond. arXiv preprint arXiv:1908.03265 (2019)

  29. Yedida, R., Saha, S., Prashanth, T.: LipschitzLR: using theoretically computed adaptive learning rates for fast convergence. arXiv preprint arXiv:1902.07399 (2019)

  30. Willmott, C.J., et al.: Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Clim. Res. 30, 79–82 (2005)

    Article  Google Scholar 

  31. Taylor, M.H., et al.: On the sensitivity of field reconstruction and prediction using empirical orthogonal functions derived from gappy data. J. Clim. 26, 9194–9205 (2013)

    Article  Google Scholar 

  32. Jerez, S., et al.: A multi-physics ensemble of present-day climate regional simulations over the Iberian Peninsula. Clim. Dyn. 40, 3023–3046 (2013). https://doi.org/10.1007/s00382-012-1539-1

    Article  Google Scholar 

  33. Ghosh, A., Kumar, H., Sastry, P.S.: Robust loss functions under label noise for deep neural networks. arXivpreprint arXiv:1712.09482 (2017)

  34. Koenker, R., Hallock, K.F.: Quantile regression. J. Econ. Perspect. 15, 143–156 (2001)

    Article  Google Scholar 

  35. Tagasovska, N., Lopez-Paz, D.: Single-model uncertainties for deep learning. In: Advances in Neural Information Processing Systems (2019)

    Google Scholar 

  36. Qi, J., et al.: On mean absolute error for deep neural network based vector-to-vector regression. IEEE Signal Process. Lett. 27, 1485–1489 (2020)

    Article  Google Scholar 

  37. Pandey, A., Wang, D.: On adversarial training and loss functions for speech enhancement. In: 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (2018)

    Google Scholar 

  38. Geurts, P., Ernst, D., Wehenkel, L.: Extremely randomized trees. Mach. Learn. 63, 3–42 (2006). https://doi.org/10.1007/s10994-006-6226-1

    Article  MATH  Google Scholar 

  39. Sen, S., Singh, K.P., Chakraborty, P:. Dealing with imbalanced regression problem for large dataset using scalable Artificial Neural Network. New Astron. 99, 101959 (2023)

    Google Scholar 

  40. Sen, S., Chakraborty, P.: A Novel Classification-Based Approach for Quicker Prediction of Redshift Using Apache Spark. In: 2022 International Conference on Data Science, Agents & Artificial Intelligence (ICDSAAI), vol. 1. IEEE (2022)

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Indian Institute of Information Technology, Allahabad, Prayagraj, India

    Snigdha Sen, Pavan Chakraborty & Krishna Pratap Singh

  2. Global Academy of Technology, Bangalore, India

    Snigdha Sen

  3. CSIS and APPCAIR BITS Pilani K K Birla Goa Campus Goa, Sancoale, India

    Snehanshu Saha

Authors
  1. Snigdha Sen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Snehanshu Saha
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Pavan Chakraborty
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Krishna Pratap Singh
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Snigdha Sen .

Editor information

Editors and Affiliations

  1. Indian Institute of Technology Indore, Indore, India

    Mohammad Tanveer

  2. Indian Institute of Information Technology - Allahabad, Prayagraj, India

    Sonali Agarwal

  3. Kobe University, Kobe, Japan

    Seiichi Ozawa

  4. Indian Institute of Technology Patna, Patna, India

    Asif Ekbal

  5. University of Innsbruck, Innsbruck, Austria

    Adam Jatowt

Rights and permissions

Reprints and permissions

Copyright information

© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Sen, S., Saha, S., Chakraborty, P., Singh, K.P. (2023). A Fast and Robust Photometric Redshift Forecasting Method Using Lipschitz Adaptive Learning Rate. In: Tanveer, M., Agarwal, S., Ozawa, S., Ekbal, A., Jatowt, A. (eds) Neural Information Processing. ICONIP 2022. Communications in Computer and Information Science, vol 1792. Springer, Singapore. https://doi.org/10.1007/978-981-99-1642-9_11

Download citation

  • DOI: https://doi.org/10.1007/978-981-99-1642-9_11

  • Published:

  • Publisher Name: Springer, Singapore

  • Print ISBN: 978-981-99-1641-2

  • Online ISBN: 978-981-99-1642-9

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics