Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Automated discovery of fundamental variables hidden in experimental data

A preprint version of the article is available at arXiv.

Abstract

All physical laws are described as mathematical relationships between state variables. These variables give a complete and non-redundant description of the relevant system. However, despite the prevalence of computing power and artificial intelligence, the process of identifying the hidden state variables themselves has resisted automation. Most data-driven methods for modelling physical phenomena still rely on the assumption that the relevant state variables are already known. A longstanding question is whether it is possible to identify state variables from only high-dimensional observational data. Here we propose a principle for determining how many state variables an observed system is likely to have, and what these variables might be. We demonstrate the effectiveness of this approach using video recordings of a variety of physical dynamical systems, ranging from elastic double pendulums to fire flames. Without any prior knowledge of the underlying physics, our algorithm discovers the intrinsic dimension of the observed dynamics and identifies candidate sets of state variables.

Your institute does not have access to this article

Access options

Buy article

Get time limited or full article access on ReadCube.

$32.00

All prices are NET prices.

Fig. 1: Two-stage modelling of dynamical systems.
Fig. 2: Prediction visualizations and physics evaluations.
Fig. 3: ID and neural state variables.
Fig. 4: Long-term prediction stability.
Fig. 5: Neural state variables for dynamics stability indicators.
Fig. 6: Neural state variables for robust long-term prediction.

Data availability

All of our simulated and physical dataset repository is available51. Source data for Figs. 2b, 3b, 4a, 5 and 6a and Extended Data Fig. 2 are available for this Article.

Code availability

The open-source code to reproduce our training and evaluation results is available at the Zenodo repository52 and GitHub (https://github.com/BoyuanChen/neural-state-variables).

References

  1. Anderson, P. W. More is different. Science 177, 393–396 (1972).

    Article  Google Scholar 

  2. Thompson, J. M. T. & Stewart, H. B. Nonlinear Dynamics and Chaos (Wiley, 2002).

  3. Hirsch, M. W., Smale, S. & Devaney, R. L. Differential Equations, Dynamical Systems, and an Introduction to Chaos (Academic, 2012).

  4. Kutz, J. N., Brunton, S. L., Brunton, B. W. & Proctor, J. L. Dynamic Mode Decomposition: Data-Driven Modeling of Complex Systems (SIAM, 2016).

  5. Evans, J. & Rzhetsky, A. Machine science. Science 329, 399–400 (2010).

    Article  Google Scholar 

  6. Fortunato, S. et al. Science of science. Science 359, eaao0185 (2018).

    Article  Google Scholar 

  7. Bongard, J. & Lipson, H. Automated reverse engineering of nonlinear dynamical systems. Proc. Natl Acad. Sci. USA 104, 9943–9948 (2007).

    Article  Google Scholar 

  8. Schmidt, M. & Lipson, H. Distilling free-form natural laws from experimental data. Science 324, 81–85 (2009).

    Article  Google Scholar 

  9. King, R. D., Muggleton, S. H., Srinivasan, A. & Sternberg, M. Structure–activity relationships derived by machine learning: the use of atoms and their bond connectivities to predict mutagenicity by inductive logic programming. Proc. Natl Acad. Sci. USA 93, 438–442 (1996).

    Article  Google Scholar 

  10. Waltz, D. & Buchanan, B. G. Automating science. Science 324, 43–44 (2009).

    Article  Google Scholar 

  11. King, R. D. et al. The robot scientist Adam. Computer 42, 46–54 (2009).

    Article  Google Scholar 

  12. Langley, P. BACON: a production system that discovers empirical laws. In Proc. Fifth International Joint Conference on Artificial Intelligence Vol. 1 344 (Morgan Kaufmann, 1977).

  13. Langley, P. Rediscovering physics with BACON.3. In Proc. Sixth International Joint Conference on Artificial Intelligence Vol. 1 505–507 (Morgan Kaufmann, 1979).

  14. Crutchfield, J. P. & McNamara, B. Equations of motion from a data series. Complex Syst. 1, 417–452 (1987).

    MathSciNet  MATH  Google Scholar 

  15. Kevrekidis, I. G. et al. Equation-free, coarse-grained multiscale computation: enabling microscopic simulators to perform system-level analysis. Commun. Math. Sci. 1, 715–762 (2003).

    MathSciNet  Article  Google Scholar 

  16. Yao, C. & Bollt, E. M. Modeling and nonlinear parameter estimation with Kronecker product representation for coupled oscillators and spatiotemporal systems. Physica D 227, 78–99 (2007).

    MathSciNet  Article  Google Scholar 

  17. Rowley, C. W., Mezić, I., Bagheri, S., Schlatter, P. & Henningson, D. S. Spectral analysis of nonlinear flows. J. Fluid Mech. 641, 115–127 (2009).

    MathSciNet  Article  Google Scholar 

  18. Schmidt, M. D. et al. Automated refinement and inference of analytical models for metabolic networks. Phys. Biol. 8, 055011 (2011).

    Article  Google Scholar 

  19. Sugihara, G. et al. Detecting causality in complex ecosystems. Science 338, 496–500 (2012).

    Article  Google Scholar 

  20. Ye, H. et al. Equation-free mechanistic ecosystem forecasting using empirical dynamic modeling. Proc. Natl Acad. Sci. USA 112, E1569–E1576 (2015).

    Google Scholar 

  21. Daniels, B. C. & Nemenman, I. Automated adaptive inference of phenomenological dynamical models. Nat. Commun. 6, 8133 (2015).

    Article  Google Scholar 

  22. Daniels, B. C. & Nemenman, I. Efficient inference of parsimonious phenomenological models of cellular dynamics using S-systems and alternating regression. PloS ONE 10, e0119821 (2015).

    Article  Google Scholar 

  23. Benner, P., Gugercin, S. & Willcox, K. A survey of projection-based model reduction methods for parametric dynamical systems. SIAM Rev. 57, 483–531 (2015).

    MathSciNet  Article  Google Scholar 

  24. Brunton, S. L., Proctor, J. L. & Kutz, J. N. Discovering governing equations from data by sparse identification of nonlinear dynamical systems. Proc. Natl Acad. Sci. USA 113, 3932–3937 (2016).

    MathSciNet  Article  Google Scholar 

  25. Rudy, S. H., Brunton, S. L., Proctor, J. L. & Kutz, J. N. Data-driven discovery of partial differential equations. Sci. Adv. 3, e1602614 (2017).

    Article  Google Scholar 

  26. Udrescu, S.-M. & Tegmark, M. AI Feynman: a physics-inspired method for symbolic regression. Sci. Adv. 6, eaay2631 (2020).

    Article  Google Scholar 

  27. Mrowca D. et al. Flexible neural representation for physics prediction. In Advances in Neural Information Processing Systems Vol. 31 (eds Bengio, S. et al.) (Curran Associates, 2018).

  28. Champion, K., Lusch, B., Kutz, J. N. & Brunton, S. L. Data-driven discovery of coordinates and governing equations. Proc. Natl Acad. Sci. USA 116, 22445–22451 (2019).

    MathSciNet  Article  Google Scholar 

  29. Baldi, P. & Hornik, K. Neural networks and principal component analysis: learning from examples without local minima. Neural Netw. 2, 53–58 (1989).

    Article  Google Scholar 

  30. Hinton, G. E. & Zemel, R. S. Autoencoders, minimum description length, and Helmholtz free energy. Adv. Neural Inf. Process. Syst. 6, 3 (1994).

    Google Scholar 

  31. Masci, J., Meier, U., Cireşan, D. & Schmidhuber, J. Stacked convolutional autoencoders for hierarchical feature extraction. In International Conference on Artificial Neural Networks 52–59 (Springer, 2011).

  32. Bishop C. M. et al. Neural Networks for Pattern Recognition (Oxford Univ. Press, 1995).

  33. Camastra, F. & Staiano, A. Intrinsic dimension estimation: advances and open problems. Inf. Sci. 328, 26–41 (2016).

    Article  Google Scholar 

  34. Campadelli, P., Casiraghi, E., Ceruti, C. & Rozza, A. Intrinsic dimension estimation: relevant techniques and a benchmark framework. Math. Probl. Eng. 2015, 759567 (2015).

  35. Levina, E. & Bickel, P. J. Maximum likelihood estimation of intrinsic dimension. In Proc. 17th International Conference on Neural Information Processing Systems 777–784 (MIT Press, 2005).

  36. Rozza, A., Lombardi, G., Ceruti, C., Casiraghi, E. & Campadelli, P. Novel high intrinsic dimensionality estimators. Mach. Learn. 89, 37–65 (2012).

    MathSciNet  Article  Google Scholar 

  37. Ceruti, C. et al. DANCo: an intrinsic dimensionality estimator exploiting angle and norm concentration. Pattern Recognit. 47, 2569–2581 (2014).

    Article  Google Scholar 

  38. Hein, M. & Audibert, J.-Y. Intrinsic dimensionality estimation of submanifolds in Rd. In Proc. 22nd International Conference on Machine Learning 289–296 (Association for Computing Machinery, 2005).

  39. Grassberger, P. & Procaccia, I. in The Theory of Chaotic Attractors 170–189 (Springer, 2004).

  40. Pukrittayakamee, A. et al. Simultaneous fitting of a potential-energy surface and its corresponding force fields using feedforward neural networks. J. Chem. Phys. 130, 134101 (2009).

    Article  Google Scholar 

  41. Wu, J., Lim, J. J., Zhang, H., Tenenbaum, J. B. & Freeman, W. T. Physics 101: Learning physical object properties from unlabeled videos. In Proc. British Machine Vision Conference (BMVC) (eds Wilson, R. C. et al.) 39.1-39.12 (BMVA Press, 2016).

  42. Chmiela, S. et al. Machine learning of accurate energy-conserving molecular force fields. Sci. Adv. 3, e1603015 (2017).

    Article  Google Scholar 

  43. Schütt, K. T., Arbabzadah, F., Chmiela, S., Müller, K. R. & Tkatchenko, A. Quantum-chemical insights from deep tensor neural networks. Nat. Commun. 8, 13890 (2017).

    Article  Google Scholar 

  44. Smith, J. S., Isayev, O. & Roitberg, A. E. ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost. Chem. Sci. 8, 3192–3203 (2017).

    Article  Google Scholar 

  45. Lutter, M., Ritter, C. & Peters, J. Deep Lagrangian networks: using physics as model prior for deep learning. In International Conference on Learning Representations (2019).

  46. Bondesan, R. & Lamacraft, A. Learning symmetries of classical integrable systems. Preprint at https://arxiv.org/abs/1906.04645 (2019).

  47. Greydanus, S. J., Dzumba, M. & Yosinski, J. Hamiltonian neural networks. Preprint at https://arxiv.org/abs/1906.01563 (2019).

  48. Swischuk, R., Kramer, B., Huang, C. & Willcox, K. Learning physics-based reduced-order models for a single-injector combustion process. AIAA J. 58, 2658–2672 (2020).

    Article  Google Scholar 

  49. Lange, H., Brunton, S. L. & Kutz, J. N. From Fourier to Koopman: spectral methods for long-term time series prediction. J. Mach. Learn. Res. 22, 1–38 (2021).

    MathSciNet  MATH  Google Scholar 

  50. Mallen, A., Lange, H. & Kutz, J. N. Deep probabilistic Koopman: long-term time-series forecasting under periodic uncertainties. Preprint at https://arxiv.org/abs/2106.06033 (2021).

  51. Chen B. et al. Dataset for the paper titled Discovering State Variables Hidden in Experimental Data (1.0). Zenodo https://doi.org/10.5281/zenodo.6653856 (2022).

  52. Chen B. et al. BoyuanChen/neural-state-variables: (v1.0). Zenodo https://doi.org/10.5281/zenodo.6629185 (2022).

Download references

Acknowledgements

This research was supported in part by NSF AI Institute for Dynamical Systems grant 2112085 (to H.L.), DARPA MTO Lifelong Learning Machines (L2M) Program HR0011-18-2-0020 (to H.L.), NSF NRI grant 1925157 (to H.L.), NSF DMS grant 1937254 (to Q.D.), NSF DMS grant 2012562 (to Q.D.), NSF CCF grant 1704833 (to Q.D.), DE grant SC0022317 (to Q.D.) and DOE ASCR DE grant SC0022317 (to Q.D.).

Author information

Authors and Affiliations

Authors

Contributions

B.C. and H.L. proposed the research; B.C., K.H., H.L. and Q.D. performed experiments and numerical analysis, B.C. and K.H. designed the algorithms; B.C., K.H., I.C. and S.R. collected the dataset; B.C., K.H., H.L. and Q.D. wrote the paper; all authors provided feedback.

Corresponding author

Correspondence to Boyuan Chen.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Computational Science thanks Bryan Daniels and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editor: Jie Pan, in collaboration with the Nature Computational Science team. Peer reviewer reports are available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 What state variables describe these dynamical systems?.

What state variables describe these dynamical systems? Identifying state variables from raw observation data is a precursor step to discovering physical laws. The key challenge is to figure out how many variables will give a complete and non-redundant description of the system’s states, what are the candidate variables, and how the variables are dependent on each other. Our work studies how to retrieve possible set of state variables from data distributions non-linearly embedded in the ambient space.

Extended Data Fig. 2 PCA and Neural State Variables visualization.

PCA and Neural State Variables visualization. Here we visualize the interesting symmetrical structures encoded in the Neural State Variables from single pendulum (A) and rigid double pendulum (B) after applying PCA algorithm on them. The colors represent the value of different physical variables. The x-axis and y-axis represent different components of the Neural State Variables.

Source data

Supplementary information

Supplementary Information

Supplementary Figs. 1–12, Discussion and Tables 1–10.

Peer review file

Supplementary Video 1

Overview video.

Supplementary Video 2

ID estimation results.

Supplementary Video 3

Long-term prediction stability results.

Supplementary Video 4

Robust prediction results for the single-pendulum system.

Supplementary Video 5

Robust prediction results for the double-pendulum system.

Source data

Source Data Fig. 3

Source data to reproduce the figure results.

Source Data Fig. 4

Source data to reproduce the figure results.

Source Data Fig. 5

Source data to reproduce the figure results.

Source Data Fig. 6

Source data to reproduce the figure results.

Source Data Extended Data Fig. 2

Source data to produce Extended Data Fig. 2.

Rights and permissions

Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Chen, B., Huang, K., Raghupathi, S. et al. Automated discovery of fundamental variables hidden in experimental data. Nat Comput Sci 2, 433–442 (2022). https://doi.org/10.1038/s43588-022-00281-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s43588-022-00281-6

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing