Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
184 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Data driven modeling for self-similar dynamics (2310.08282v3)

Published 12 Oct 2023 in cs.LG and cond-mat.stat-mech

Abstract: Multiscale modeling of complex systems is crucial for understanding their intricacies. Data-driven multiscale modeling has emerged as a promising approach to tackle challenges associated with complex systems. On the other hand, self-similarity is prevalent in complex systems, hinting that large-scale complex systems can be modeled at a reduced cost. In this paper, we introduce a multiscale neural network framework that incorporates self-similarity as prior knowledge, facilitating the modeling of self-similar dynamical systems. For deterministic dynamics, our framework can discern whether the dynamics are self-similar. For uncertain dynamics, it can compare and determine which parameter set is closer to self-similarity. The framework allows us to extract scale-invariant kernels from the dynamics for modeling at any scale. Moreover, our method can identify the power law exponents in self-similar systems. Preliminary tests on the Ising model yielded critical exponents consistent with theoretical expectations, providing valuable insights for addressing critical phase transitions in non-equilibrium systems.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (65)
  1. F. Taubert, R. Fischer, J. Groeneveld, S. Lehmann, M. S. Müller, E. Rödig, T. Wiegand,  and A. Huth, “Global patterns of tropical forest fragmentation,” Nature 554, 519–522 (2018).
  2. O. Peters and J. D. Neelin, “Critical phenomena in atmospheric precipitation,” Nature Physics 2, 393–396 (2006).
  3. R. D. Smith, “The dynamics of internet traffic: self-similarity, self-organization, and complex phenomena,” Advances in Complex Systems 14, 905–949 (2011).
  4. D. Notarmuzi, C. Castellano, A. Flammini, D. Mazzilli,  and F. Radicchi, “Universality, criticality and complexity of information propagation in social media,” Nature Communications , 1–8 (2021).
  5. S. Herculano-Houzel, P. R. Manger,  and J. H. Kaas, “Brain scaling in mammalian evolution as a consequence of concerted and mosaic changes in numbers of neurons and average neuronal cell size,” Frontiers in Neuroanatomy 8, 1–28 (2014).
  6. D. La Rocca, N. Zilber, P. Abry, V. van Wassenhove,  and P. Ciuciu, “Self-similarity and multifractality in human brain activity: A wavelet-based analysis of scale-free brain dynamics,” Journal of Neuroscience Methods 309, 175–187 (2018).
  7. T. L. Ribeiro, D. R. Chialvo,  and D. Plenz, “Scale-Free Dynamics in Animal Groups and Brain Networks,” Frontiers in Systems Neuroscience 14, 1–10 (2021).
  8. G. F. Grosu, A. V. Hopp, V. V. Moca, H. Bârzan, A. Ciuparu, M. Ercsey-Ravasz, M. Winkel, H. Linde,  and R. C. Mureşan, “The fractal brain: scale-invariance in structure and dynamics,” Cerebral Cortex 33, 4574–4605 (2023).
  9. C. Song and S. Havlin, “Self-similarity of complex networks,” Nature 433, 2–5 (2005).
  10. F. Radicchi, J. J. Ramasco, A. Barrat,  and S. Fortunato, “Complex networks renormalization: Flows and fixed points,” Physical Review Letters 101, 3–6 (2008).
  11. F. Radicchi, A. Barrat, S. Fortunato,  and J. J. Ramasco, “Renormalization flows in complex networks,” Physical Review E 79, 1–11 (2009).
  12. H. D. Rozenfeld, C. Song,  and H. A. Makse, “Small-world to fractal transition in complex networks: A renormalization group approach,” Physical Review Letters 104, 1–4 (2010).
  13. G. García-Pérez, M. Boguñá,  and M. Á. Serrano, “Multiscale unfolding of real networks by geometric renormalization,” Nature Physics 14, 583–589 (2018).
  14. D. Chen, H. Su, X. Wang, G. J. Pan,  and G. Chen, “Finite-size scaling of geometric renormalization flows in complex networks,” Physical Review E 104, 1–12 (2021).
  15. D. Chen, H. Su,  and Z. Zeng, “Geometric Renormalization Reveals the Self-Similarity of Weighted Networks,” IEEE Transactions on Computational Social Systems 10, 426–434 (2023).
  16. P. Villegas, T. Gili, G. Caldarelli,  and A. Gabrielli, “Laplacian renormalization group for heterogeneous networks,” Nature Physics 19, 445–450 (2023).
  17. A. C. Antoulas, “Approximation of large-scale dynamical systems: An overview,” IFAC Proceedings Volumes (IFAC-PapersOnline) 37, 19–28 (2004).
  18. D. J. Lucia, P. S. Beran,  and W. A. Silva, “Reduced-order modeling: new approaches for computational physics,” Progress in Aerospace Sciences 40, 51–117 (2004).
  19. D. M. Luchtenburg, “Data-driven science and engineering: machine learning, dynamical systems, and control,” IEEE Control Systems Magazine 41, 95–102 (2021).
  20. P. J. Schmid, “Dynamic mode decomposition and its variants,” Annual Review of Fluid Mechanics 54, 225–254 (2022).
  21. S. L. Brunton, M. Budišić, E. Kaiser,  and J. N. Kutz, “Modern koopman theory for dynamical systems,” arXiv preprint arXiv:2102.12086  (2021).
  22. I. G. Kevrekidis, C. W. Gear, J. M. Hyman, P. G. Kevrekidis, O. Runborg, C. Theodoropoulos, et al., “Equation-free, coarse-grained multiscale computation: enabling microscopic simulators to perform system-level analysis,” Commun. Math. Sci 1, 715–762 (2003).
  23. I. G. Kevrekidis, C. W. Gear,  and G. Hummer, “Equation-free: The computer-aided analysis of complex multiscale systems,” AIChE Journal 50, 1346–1355 (2004).
  24. M. O. Williams, I. G. Kevrekidis,  and C. W. Rowley, “A data-driven approximation of the koopman operator: Extending dynamic mode decomposition,” Journal of Nonlinear Science 25, 1307–1346 (2015).
  25. L. P. Kadanoff, “Scaling laws for ising models near tc,” Physics Physique Fizika 2, 263 (1966).
  26. K. G. Wilson, “The renormalization group and critical phenomena,” Reviews of Modern Physics 55, 583 (1983).
  27. A. Pelissetto and E. Vicari, “Critical phenomena and renormalization-group theory,” Physics Reports 368, 549–727 (2002).
  28. U. Schollwöck, “The density-matrix renormalization group,” Reviews of modern physics 77, 259 (2005).
  29. Tauber, critical dynamics (2014).
  30. A. Cavagna, D. Conti, C. Creato, L. Del Castello, I. Giardina, T. S. Grigera, S. Melillo, L. Parisi,  and M. Viale, “Dynamic scaling in natural swarms,” Nature Physics 13, 914–918 (2017).
  31. M. C. Yalabik and J. D. Gunton, “Monte Carlo renornnlization-group studies of kinetic Ising models,” Physical Review B 25, 2–5 (1982).
  32. G. F. Mazenko, J. E. Hirsch,  and J. Nolan, “Real-space dynamic renormalization group. III. Calculation of correlation functions,”  23, 1431–1446 (1981).
  33. A. Vespignani, S. Zapperi,  and P. Luciano, “Renormalization approach to the self-organized critical behavior of sandpile models,” Physical Review E 2, 243–253 (1995).
  34. E. V. Ivashkevich, A. M. Povolotsky, A. Vespignani,  and S. Zapperi, “Dynamical real space renormalization group applied to sandpile models,” Physical Review E 60, 1239–1251 (1999).
  35. C. Y. Lin and C. K. Hu, “Renormalization-group approach to an Abelian sandpile model on planar lattices,” Physical Review E 66, 1–29 (2002).
  36. A. Cavagna, L. Di Carlo, I. Giardina, L. Grandinetti, T. S. Grigera,  and G. Pisegna, “Dynamical Renormalization Group Approach to the Collective Behavior of Swarms,” Physical Review Letters 123, 268001 (2019a).
  37. A. Cavagna, L. Di Carlo, I. Giardina, L. Grandinetti, T. S. Grigera,  and G. Pisegna, “Renormalization group crossover in the critical dynamics of field theories with mode coupling terms,” Physical Review E 100, 1–29 (2019b).
  38. A. Cavagna, P. M. Chaikin, D. Levine, S. Martiniani, A. Puglisi,  and M. Viale, “Vicsek model by time-interlaced compression: A dynamical computable information density,” Physical Review E 103, 1–10 (2021).
  39. I. Moise and R. Temam, “Renormalization group method: Application to navier-stokes equation,” Discrete and Continuous Dynamical Systems 6, 191–210 (1999).
  40. N. Israeli and N. Goldenfeld, “Coarse-graining of cellular automata, emergence, and the predictability of complex systems,” Physical Review E 73, 1–17 (2006).
  41. G. E. Karniadakis, I. G. Kevrekidis, L. Lu, P. Perdikaris, S. Wang,  and L. Yang, “Physics-informed machine learning,” Nature Reviews Physics 3, 422–440 (2021).
  42. L. G. Wright, T. Onodera, M. M. Stein, T. Wang, D. T. Schachter, Z. Hu,  and P. L. McMahon, “Deep physical neural networks trained with backpropagation,” Nature 601, 549–555 (2022).
  43. R. Vinuesa and S. L. Brunton, “Enhancing computational fluid dynamics with machine learning,” Nature Computational Science 2, 358–366 (2022).
  44. M. Raissi, P. Perdikaris,  and G. E. Karniadakis, “Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations,” Journal of Computational physics 378, 686–707 (2019).
  45. B. Lusch, J. N. Kutz,  and S. L. Brunton, “Deep learning for universal linear embeddings of nonlinear dynamics,” Nature communications 9, 4950 (2018).
  46. O. Azencot, N. B. Erichson, V. Lin,  and M. Mahoney, “Forecasting sequential data using consistent koopman autoencoders,” in International Conference on Machine Learning (PMLR, 2020) pp. 475–485.
  47. P. Bevanda, S. Sosnowski,  and S. Hirche, “Koopman operator dynamical models: Learning, analysis and control,” Annual Reviews in Control 52, 197–212 (2021).
  48. S. E. Otto and C. W. Rowley, “Koopman operators for estimation and control of dynamical systems,” Annual Review of Control, Robotics, and Autonomous Systems 4, 59–87 (2021).
  49. S. L. Brunton, B. R. Noack,  and P. Koumoutsakos, “Machine learning for fluid mechanics,” Annual review of fluid mechanics 52, 477–508 (2020).
  50. P. R. Vlachas, W. Byeon, Z. Y. Wan, T. P. Sapsis,  and P. Koumoutsakos, “Data-driven forecasting of high-dimensional chaotic systems with long short-term memory networks,” Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences 474, 20170844 (2018).
  51. P. R. Vlachas, G. Arampatzis, C. Uhler,  and P. Koumoutsakos, “Multiscale simulations of complex systems by learning their effective dynamics,” Nature Machine Intelligence 4, 359–366 (2022), arXiv:2006.13431 .
  52. L. Feng, T. Gao, M. Dai,  and J. Duan, “Learning effective dynamics from data-driven stochastic systems,” Chaos: An Interdisciplinary Journal of Nonlinear Science 33 (2023).
  53. N. Walker, K. M. Tam,  and M. Jarrell, “Deep learning on the 2-dimensional Ising model to extract the crossover region with a variational autoencoder,” Scientific Reports 10, 1–12 (2020).
  54. K. Shiina, H. Mori, Y. Tomita, H. K. Lee,  and Y. Okabe, “Inverse renormalization group based on image super-resolution using deep convolutional networks,” Scientific Reports 11, 1–9 (2021).
  55. H. Y. Hu, S. H. Li, L. Wang,  and Y. Z. You, “Machine learning holographic mapping by neural network renormalization group,” Physical Review Research 2, 23369 (2020).
  56. W. Hou and Y.-Z. You, “Machine Learning Renormalization Group for Statistical Physics,” arXiv preprint , 1–13 (2023), arXiv:2306.11054 .
  57. D. Ron, A. Brandt,  and R. H. Swendsen, “Monte Carlo renormalization-group calculation for the d=3 Ising model using a modified transformation,” Physical Review E 104 (2021).
  58. J. H. Chung and Y. J. Kao, “Neural Monte Carlo renormalization group,” Physical Review Research 3 (2021).
  59. S. H. Li and L. Wang, “Neural Network Renormalization Group,” Physical Review Letters 121, 1–10 (2018).
  60. P. M. Lenggenhager, D. E. Gökmen, Z. Ringel, S. D. Huber,  and M. Koch-Janusz, “Optimal renormalization group transformation from information theory,” Physical Review X 10, 11037 (2020), 1809.09632 .
  61. S. Iso, S. Shiba,  and S. Yokoo, “Scale-invariant feature extraction of neural network and renormalization group flow,” Physical Review E 97, 1–16 (2018).
  62. J. Zhang and K. Liu, “Neural information squeezer for causal emergence,” Entropy 25, 26 (2022).
  63. T. Vicsek, A. Czirk, E. Ben-Jacob, I. Cohen,  and O. Shochet, “Novel type of phase transition in a system of self-driven particles,” Physical Review Letters 75, 1226–1229 (1995), arXiv:0611743 [cond-mat] .
  64. E. P. Hoel, L. Albantakis,  and G. Tononi, “Quantifying causal emergence shows that macro can beat micro,” Proceedings of the National Academy of Sciences of the United States of America 110, 19790–19795 (2013).
  65. M. Yang, Z. Wang, K. Liu, Y. Rong, B. Yuan,  and J. Zhang, “Finding emergence in data by maximizing effective information,”  (2023), arXiv:2308.09952 [physics.soc-ph] .

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now