Physics-Informed Classical Lagrange / Hamilton Neural Networks in Deep Learning
Abstract
The principles of constructing deep machine learning systems based on taking into account information about the physical properties of the studied control object, such as an autonomous robot, are considered. The platform for the development of intelligent tools is deep machine learning models applying physics-informed neural networks. Most of the methods under development for constructing system identification models are either "black box" models (i.e., general models based on training data) or so-called "white box" models (e.g., state-space/control models, which can be explicitly expressed mathematically). Thus, the direction of development is to study the gray box model in the state space. A gray box model means a model that is trained on data while being guided by information about some physical properties or laws that apply. Such models can be further applied for adaptive сontrol and self-organization. Therefore, the use of state-space models is considered. A feature of physically informed neural networks is that they initially take into account the underlying description of the physical interpretation of partial or ordinary differential equations, that is, the physics of the problem, instead of trying to derive a solution based solely on data, that is, by approximating a set of pairs by a neural network "state-value". Lagrange and Newton neural networks are considered as such a class of learning systems. Specific examples show the advantages and features of using the discussed types of physics-informed neural networks.
References
2. Thuerey N., et al. Physics-based Deep Learning. arXiv:2109.05237. 2022. Available at: https://arxiv.org/pdf/2109.05237.pdf (accessed 19.05.2022). (In Eng.)
3. Raissi M., Perdikaris P., Karniadakis G.E. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational Physics. 2019; 378:686-707. (In Eng.) doi: https://doi.org/10.1016/j.jcp.2018.10.045
4. Karniadakis G.E., Kevrekidis I.G., Lu L., et al. Physics-informed machine learning. Nature Reviews Physics. 2021; 3:423-440. (In Eng.) doi: http://dx.doi.org/10.1038/s42254-021-00314-5
5. Cuomo S. et al. Scientific Machine Learning Through Physics-Informed Neural Networks: Where we are and What’s Next. Journal of Scientific Computing. 2022; 92(3):88. (In Eng.) doi: https://doi.org/10.48550/arXiv.2201.05624
6. Greydanus S., Dzamba M., Yosinski J. Hamiltonian Neural Networks. In: Wallach H., Larochelle H., Beygelzimer A., d'Alché-Buc F., Fox E., Garnett R. (eds.) Advances in Neural Information Processing Systems (NeurIPS 2019). Vol. 32. Curran Associates, Inc.; 2019. p. 1-11. Available at: https://proceedings.neurips.cc/paper/2019/file/26cd8ecadce0d4efd6cc8a8725cbd1f8-Paper.pdf (accessed 19.05.2022). (In Eng.)
7. Sosanya A., Greydanus S. Dissipative Hamiltonian Neural Networks: Learning Dissipative and Conservative Dynamics Separately. arXiv:2201.10085. 2022. Available at: https://arxiv.org/pdf/2201.10085.pdf (accessed 19.05.2022). (In Eng.)
8. Dhulipala S.L.N., Che Y., Shields M.D. Bayesian Inference with Latent Hamiltonian Neural Networks. arXiv:2208.06120. 2022. (In Eng.) doi: https://doi.org/10.48550/arXiv.2208.06120
9. Cranmer M., Greydanus S., Hoyer S., Battaglia P., Spergel D., Ho S. Lagrangian Neural Networks. ICLR 2020 Workshop on Integration of Deep Neural Models and Differential Equations. Addis Ababa, Ethiopia; 2019. Available at: https://openreview.net/forum?id=iE8tFa4Nq (accessed 19.05.2022). (In Eng.)
10. Liu Z., Wang B., Meng Q., Chen W., Tegmark M., Liu T.-Y. Machine-Learning Non-Conservative Dynamics for New-Physics Detection. Physical Review E. 2021; 104:055302. (In Eng.) doi: https://doi.org/10.1103/PhysRevE.104.055302
11. Meng C., Seo S., Cao D., Griesemer S., Liu Y. When Physics Meets Machine Learning: A Survey of Physics-Informed Machine Learning. arXiv:2203.16797. 2022. (In Eng.) doi: https://doi.org/10.48550/arXiv.2203.16797
12. Zhai H., Sands T. Controlling Chaos in Van Der Pol Dynamics Using Signal-Encoded Deep Learning. Mathematics. 2022; 10(3):453. (In Eng.) doi: https://doi.org/10.3390/math10030453
13. Denman H.H., Buch L.H. Solution of the Hamilton-Jacobi equation for certain dissipative classical mechanical systems. Journal of Mathematical Physics. 1973; 14:326-329. (In Eng.) doi: https://doi.org/10.1063/1.1666316
14. Ohsawa T., Bloch A.M. Nonholomonic Hamilton-Jacobi equation and integrability. Journal of Geometric Mechanics. 2009; 1(4):461-481. (In Eng.) doi: https://doi.org/10.3934/jgm.2009.1.461
15. Balseiro P., Marrero J.C., de Diego D.M., Padron E. A unified framework for mechanics: Hamilton-Jacobi equation and applications. Nonlinearity. 2010; 23(8):1887-1918. (In Eng.) doi: https://doi.org/10.1088/0951-7715/23/8/006
16. Flannery M.R. d’Alembert-Lagrange analytical dynamics for nonholonomic systems. Journal of Mathematical Physics. 2011; 52(3):032705. (In Eng.) doi: https://doi.org/10.1063/1.3559128
17. Litvintseva L.V., Ul’yanov S.V., Ul’yanov S.S. Design of robust knowledge bases of fuzzy controllers for intelligent control of substantially nonlinear dynamic systems: II. A soft computing optimizer and robustness of intelligent control systems. Journal of Computer and Systems Sciences International. 2006; 45(5):744-771. (In Eng.) doi: https://doi.org/10.1134/S106423070605008X
18. Valelis C., Anagnostopoulos F.K., Basilakos S., Saridakis E.N. Building healthy Lagrangian theories with machine learning. International Journal of Modern Physics D. 2021; 30(11):2150085. (In Eng.) doi: https://doi.org/10.1142/S0218271821500851
19. Waltz D., Buchanan B.G. Automating Science. Science. 2009; 324(5923):43-44. (In Eng.) doi: https://doi.org/10.1126/science.1172781
20. Chiu P.-H., Wong J.C., Ooi C., Dao M.H., Ong Y.-S. CAN-PINN: A fast physics-informed neural network based on coupled-automatic-numerical differentiation method. Computer Methods in Applied Mechanics and Engineering. 2022; 395:114909. (In Eng.) doi: https://doi.org/10.1016/j.cma.2022.114909
21. Fang Z. A High-Efficient Hybrid Physics-Informed Neural Networks Based on Convolutional Neural Network. IEEE Transactions on Neural Networks and Learning Systems. 2022; 33(10):5514-5526. (In Eng.) doi: https://doi.org/10.1109/TNNLS.2021.3070878
22. Rouhani B.D., Mirhoseini A., Koushanfar F. Going deeper than deep learning for massive data analytics under physical constraints. Proceedings of the Eleventh IEEE/ACM/IFIP International Conference on Hardware/Software Codesign and System Synthesis (CODES '16). Association for Computing Machinery, New York, NY, USA; 2016. Article number: 17. p. 1-3. (In Eng.) doi: https://doi.org/10.1145/2968456.2976766
23. Zhang Y. Towards Piecewise-Linear Primal Neural Networks for Optimization and Redundant Robotics. 2006 IEEE International Conference on Networking, Sensing and Control. IEEE Computer Society; 2006. p. 374-379. (In Eng.) doi: https://doi.org/10.1109/ICNSC.2006.1673175
24. Zhang C., Shafieezadeh A. Simulation-free reliability analysis with active learning and Physics-Informed Neural Network. Reliability Engineering & System Safety. 2022; 226:108716. (In Eng.) doi: https://doi.org/10.1016/j.ress.2022.108716
25. Dierkes E., Flaßkamp K. Learning Hamiltonian Systems considering System Symmetries in Neural Networks. IFAC-PapersOnLine. 2021; 54(19):210-216. (In Eng.) doi: https://doi.org/10.1016/j.ifacol.2021.11.080
This work is licensed under a Creative Commons Attribution 4.0 International License.
Publication policy of the journal is based on traditional ethical principles of the Russian scientific periodicals and is built in terms of ethical norms of editors and publishers work stated in Code of Conduct and Best Practice Guidelines for Journal Editors and Code of Conduct for Journal Publishers, developed by the Committee on Publication Ethics (COPE). In the course of publishing editorial board of the journal is led by international rules for copyright protection, statutory regulations of the Russian Federation as well as international standards of publishing.
Authors publishing articles in this journal agree to the following: They retain copyright and grant the journal right of first publication of the work, which is automatically licensed under the Creative Commons Attribution License (CC BY license). Users can use, reuse and build upon the material published in this journal provided that such uses are fully attributed.
