Applications of domain adversarial neural network in phase transition of 3D Potts model. (English) Zbl 1536.82006
Summary: Machine learning techniques exhibit significant performance in discriminating different phases of matter and provide a new avenue for studying phase transitions. We investigate the phase transitions of three dimensional \(q\)-state Potts model on cubic lattice by using a transfer learning approach, Domain Adversarial Neural Network (DANN). With the unique neural network architecture, it could evaluate the high-temperature (disordered) and low-temperature (ordered) phases, and identify the first and second order phase transitions. Meanwhile, by training the DANN with a few labeled configurations, the critical points for \(q = 2\), 3, 4 and 5 can be predicted with high accuracy, which are consistent with those of the Monte Carlo simulations. These findings would promote us to learn and explore the properties of phase transitions in high-dimensional systems.
MSC:
| 82B20 | Lattice systems (Ising, dimer, Potts, etc.) and systems on graphs arising in equilibrium statistical mechanics |
| 68T05 | Learning and adaptive systems in artificial intelligence |
Keywords:
machine learning; domain adversarial neural network; 3D Potts model; phase transitions; critical phenomenaReferences:
| [1] | Phase Transitions and Critical Phenomena, 2000, Elsevier · Zbl 1063.82001 |
| [2] | A Modern Course in Statistical Physics, 2016, John Wiley & Sons, https://onlinelibrary.wiley.com/doi/book/10.1002/9783527690497 · Zbl 1334.00085 |
| [3] | Statistical Mechanics of Phase Transitions, 1992, Clarendon Press |
| [4] | Statistical Physics: Volume 5, Vol. 5, 2013, Elsevier |
| [5] | Deep potential molecular dynamics: A scalable model with the accuracy of quantum mechanics, Phys. Rev. Lett., 120, 14, Article 143001 pp., 2018 |
| [6] | Learning thermodynamics with Boltzmann machines, Phys. Rev. B, 94, 16, Article 165134 pp., 2016 |
| [7] | Quantum loop topography for machine learning, Phys. Rev. Lett., 118, 21, Article 216401 pp., 2017 |
| [8] | Jet tagging algorithm of graph network with Haar pooling message passing, Phys. Rev. D, 108, 7, Article 072007 pp., 2023 |
| [9] | EAZY: A fast, public photometric redshift code, Astrophys. J., 686, 2, 1503, 2008 |
| [10] | CMU DeepLens: Deep learning for automatic image-based galaxy-galaxy strong lens finding, Mon. Not. R. Astron. Soc., 473, 3, 3895-3906, 2018 |
| [11] | Estimating cosmological parameters from the dark matter distribution, (2016, PMLR), 2407-2416 |
| [12] | Provably efficient machine learning for quantum many-body problems, Science, 377, 6613, eabk3333, 2022 · Zbl 1524.81003 |
| [13] | Machine learning quantum phases of matter beyond the Fermion sign problem, Sci. Rep., 7, 1, 1-10, 2017 |
| [14] | Machine learning for optical fiber communication systems: An introduction and overview, APL Photon., 6, 12, 2021 |
| [15] | Advancing theoretical understanding and practical performance of signal processing for nonlinear optical communications through machine learning, Nature Commun., 11, 1, 3694, 2020 |
| [16] | A machine learning-based detection technique for optical fiber nonlinearity mitigation, IEEE Photonics Technol. Lett., 31, 8, 627-630, 2019 |
| [17] | Machine learning phases of matter, Nat. Phys., 13, 5, 431-434, 2017 |
| [18] | Detection of phase transition via convolutional neural networks, J. Phys. Soc. Japan, 86, 6, Article 063001 pp., 2017 |
| [19] | Machine-learning study using improved correlation configuration and application to quantum Monte Carlo simulation, Phys. Rev. E, 102, 2, Article 021302 pp., 2020 |
| [20] | Discovering phase transitions with unsupervised learning, Phys. Rev. B, 94, 19, Article 195105 pp., 2016 |
| [21] | Dynamics of spontaneous symmetry breaking in the Weinberg-Salam theory, Phys. Rev. D, 20, 10, 2619, 1979 |
| [22] | Potts model at the critical temperature, J. Phys. C: Solid State Phys., 6, 23, L445, 1973 |
| [23] | Exactly Solved Models in Statistical Mechanics, 2016, Elsevier |
| [24] | Calculation of the critical temperature for 2- and 3-dimensional Ising models and for 2-dimensional Potts models using the transfer matrix method, Physics, 105, 42, 10355-10359, 2004, https://pubs.acs.org/doi/10.1021/jp010400o |
| [25] | Detecting signals of weakly first-order phase transitions in two-dimensional Potts models, J. Phys. Soc. Japan, 88, 3, Article 034006 pp., 2019 |
| [26] | Applications of neural networks to the studies of phase transitions of two-dimensional Potts models, Ann. Phys., 391, 312-331, 2018 |
| [27] | On the generalizability of artificial neural networks in spin models, SciPost Phys. Core, 5, 2, 032, 2022 |
| [28] | A universal neural network for learning phases, Eur. Phys. J. Plus, 136, 11, 1116, 2021 |
| [29] | A comprehensive neural networks study of the phase transitions of potts model, New J. Phys., 22, 6, Article 063016 pp., 2020 |
| [30] | Neural network flows of low q-state potts and clock models, New J. Phys., 24, 4, Article 043040 pp., 2022 |
| [31] | Phase transition properties of 3D Potts models, Nuclear Phys. B, 802, 3, 421-434, 2008 · Zbl 1190.82017 |
| [32] | Calculation of partition functions by measuring component distributions, Phys. Rev. Lett., 94, 5, Article 050601 pp., 2005 |
| [33] | Machine learning for quantum matter, Adv. Phys.: X, 5, 1, Article 1797528 pp., 2020 |
| [34] | Domain-adversarial neural networks, 2014, arXiv preprint arXiv:1412.4446 |
| [35] | Domain-adversarial training of neural networks, J. Mach. Learn. Res., 17, 59, 1-35, 2016 · Zbl 1360.68671 |
| [36] | A brief review of domain adaptation, Adv. Data Sci. Inform. Eng., 877-894, 2021 |
| [37] | Identifying quantum phase transitions with adversarial neural networks, Phys. Rev. B, 97, 13, Article 134109 pp., 2018 |
| [38] | Transfer learning of phase transitions in percolation and directed percolation, Phys. Rev. E, 105, Article 064139 pp., 2022 |
| [39] | Study of phase transition of potts model with domain adversarial neural network, Physica A, 617, Article 128666 pp., 2023 · Zbl 1521.82034 |
| [40] | Phase transitions of ferromagnetic Potts models on the simple cubic lattice, Chin. Phys. Lett., 31, 7, Article 070503 pp., 2014 |
| [41] | 3D bond-diluted 4-state Potts model: A Monte Carlo study, Nuclear Phys. B Proc. Suppl., 106, 899-901, 2002 |
| [42] | Nonuniversal critical dynamics in Monte Carlo simulations, Phys. Rev. Lett., 58, 2, 86, 1987 |
| [43] | Collective Monte Carlo updating for spin systems, Phys. Rev. Lett., 62, 4, 361, 1989 |
| [44] | Non-Equilibrium Phase Transitions. Vol. 1, 2008, Springer · Zbl 1165.82002 |
| [45] | Finite-size scaling in phase transitions and critical phenomena, Phase Transit. Crit. Phenomena, 8, 146, 1983 |
| [46] | Scaling theory for finite-size effects in the critical region, Phys. Rev. Lett., 28, 23, 1516, 1972 |
| [47] | Finite Size Scaling and Numerical Simulation of Statistical Systems, 1990, World Scientific |
| [48] | Monte Carlo study of phase transitions in the bond-diluted 3D 4-state Potts model, Nuclear Phys. B, 719, 3, 275-311, 2005 · Zbl 1207.82005 |
| [49] | An alternative order parameter for the 4-state Potts model, Physica A, 366, 255-264, 2006 |
| [50] | Conformal Invariance and Critical Phenomena, 1999, Springer Science & Business Media · Zbl 1053.82001 |
| [51] | Phase Transitions and Critical Phenomena. Vol. 7, 1971, Clarendon Press: Clarendon Press Oxford |
| [52] | Equilibrium Statistical Physics, 1994, World scientific · Zbl 0837.00008 |
| [53] | Machine learning and the physical sciences, Rev. Modern Phys., 91, 4, Article 045002 pp., 2019 |
| [54] | Solving the quantum many-body problem with artificial neural networks, Science, 355, 6325, 602-606, 2017 · Zbl 1404.81313 |
| [55] | Jet substructure classification in high-energy physics with deep neural networks, Phys. Rev. D, 93, 9, Article 094034 pp., 2016 |
| [56] | Learning to classify from impure samples with high-dimensional data, Phys. Rev. D, 98, 1, Article 011502 pp., 2018 |
| [57] | Deep autoregressive models for the efficient variational simulation of many-body quantum systems, Phys. Rev. Lett., 124, 2, Article 020503 pp., 2020 |
| [58] | Projected regression method for solving Fredholm integral equations arising in the analytic continuation problem of quantum physics, Inverse Problems, 33, 11, Article 115007 pp., 2017 · Zbl 1379.65097 |
| [59] | Learning phase transitions by confusion, Nat. Phys., 13, 5, 435-439, 2017 |
| [60] | Machine learning of phase transitions in the percolation and XY models, Phys. Rev. E, 99, 3, Article 032142 pp., 2019 |
| [61] | Some generalized order-disorder transformations, (1952, Cambridge University Press), 106-109 · Zbl 0048.45601 |
| [62] | Antiferromagnetic Potts models, Phys. Rev. Lett., 63, 2, 109, 1989 |
| [63] | Complete density calculations of q-state potts and clock models: Reentrance of interface densities under symmetry breaking, Phys. Rev. E, 102, 6, Article 062135 pp., 2020 |
| [64] | The Potts model, Rev. Modern Phys., 54, 1, 235, 1982 |
| [65] | A relation between the temperature exponents of the eight-vertex and q-state potts model, J. Phys. A: Math. Gen., 12, 10, 1857, 1979 |
| [66] | Monte Carlo study of the potts model on the square and the simple cubic lattices, Phys. Rev. B, 40, 7, 5007, 1989 |
| [67] | First-order phase transitions and the three-state potts model, Phys. Rev. Lett., 43, 11, 799, 1979 |
| [68] | Latent heat calculation of the three-dimensional q=3, 4, and 5 potts models by the tensor product variational approach, Phys. Rev. E, 65, 4, Article 046702 pp., 2002 |
| [69] | Tensor renormalization of three-dimensional Potts model, 2022, arxiv |
| [70] | Correlation length of the three-state Potts model in three dimensions, Phys. Rev. Lett., 63, 1, 13, 1989 |
| [71] | Three-dimensional q-state Potts model: Monte Carlo study near q=3, Phys. Rev. B, 43, 1, 1268, 1991 |
| [72] | A survey on transfer learning, IEEE Trans. Knowl. Data Eng., 22, 10, 1345-1359, 2009 |
| [73] | Semi-Supervised Learning Literature Survey, 2005, University of Wisconsin-Madison Department of Computer Sciences |
| [74] | A survey of transfer learning, J. Big data, 3, 1, 1-40, 2016 |
| [75] | Unsupervised machine learning account of magnetic transitions in the Hubbard model, Phys. Rev. E, 97, 1, Article 013306 pp., 2018 |
| [76] | Comparative study of damage spreading in the Ising model using heat-bath, glauber, and metropolis dynamics, J. Stat. Phys., 59, 3, 1043-1050, 1990 |
| [77] | Computational Studies of Pure and Dilute Spin Models, 2000, School of Mathematics and Computing, University of Derby |
| [78] | Time-dependent statistics of the Ising model, J. Math. Phys., 4, 2, 294-307, 1963 · Zbl 0145.24003 |
| [79] | The magnetization of the 3D Ising model, J. Phys. A: Math. Gen., 29, 17, 5727, 1996 · Zbl 0900.82010 |
| [80] | 92 Years of the Ising model: A high resolution Monte Carlo study, (2018, IOP Publishing), Article 012002 pp. |
| [81] | Normalized entropy density of the 3D 3-state Potts model, Phys. Rev. D, 75, 9, Article 094506 pp., 2007 |
| [82] | Phase diagram of QCD in a magnetic background, Phys. Rev. D, 105, 3, Article 034511 pp., 2022 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
