Three-dimensional ResNet for efficient prediction of ground state phases in multicomponent dipolar spinor BECs

doi:10.1088/1674-1056/adc7f9

中国物理B ›› 2025, Vol. 34 ›› Issue (7): 76701-076701.doi: 10.1088/1674-1056/adc7f9

Three-dimensional ResNet for efficient prediction of ground state phases in multicomponent dipolar spinor BECs

Chengji Liao(廖承继)¹, Tiantian Li(李甜甜)^1,†, Xiao-Dong Bai(柏小东)², and Yunbo Zhang(张云波)³

1 School of Physics and Optoelectronics, Xiangtan University, Xiangtan 411100, China;
2 College of Physics and Hebei Key Laboratory of Photophysics Research and Application, Hebei Normal University, Shijiazhuang 050024, China;
3 Department of Physics, Zhejiang Sci-Tech University, Hangzhou 310018, China

收稿日期:2025-02-24 修回日期:2025-03-28 接受日期:2025-04-02 出版日期:2025-06-18 发布日期:2025-07-10
通讯作者: Tiantian Li E-mail:ttli@xtu.edu.cn
基金资助:
Project supported by the National Natural Science Foundation of China (Grant Nos. 11904309 and 12305015), the Natural Science Foundation of Hunan Province, China (Grant No. 2020JJ5528), and the Natural Science Foundation of Hebei Province, China (Grant No. A2024205027).

Three-dimensional ResNet for efficient prediction of ground state phases in multicomponent dipolar spinor BECs

Chengji Liao(廖承继)¹, Tiantian Li(李甜甜)^1,†, Xiao-Dong Bai(柏小东)², and Yunbo Zhang(张云波)³

1 School of Physics and Optoelectronics, Xiangtan University, Xiangtan 411100, China;
2 College of Physics and Hebei Key Laboratory of Photophysics Research and Application, Hebei Normal University, Shijiazhuang 050024, China;
3 Department of Physics, Zhejiang Sci-Tech University, Hangzhou 310018, China

Received:2025-02-24 Revised:2025-03-28 Accepted:2025-04-02 Online:2025-06-18 Published:2025-07-10
Contact: Tiantian Li E-mail:ttli@xtu.edu.cn
Supported by:
Project supported by the National Natural Science Foundation of China (Grant Nos. 11904309 and 12305015), the Natural Science Foundation of Hunan Province, China (Grant No. 2020JJ5528), and the Natural Science Foundation of Hebei Province, China (Grant No. A2024205027).

摘要/Abstract

摘要： Machine learning has rapidly become a powerful tool for addressing challenges in ultracold atomic systems; however, its application to intricate three-dimensional (3D) systems remains relatively underexplored. In this study, we introduce a 3D residual network (3D ResNet) framework based on 3D convolutional neural networks (3D CNN) to predict ground states phases in 3D dipolar spinor Bose-Einstein condensates (BECs). Our results show that the 3D ResNet framework predicts ground states with high accuracy and efficiency across a broad parameter space. To enhance phase transition predictions, we incorporate data augmentation techniques, leading to a notable improvement in the model's performance. The method is further validated in more complex scenarios, particularly when transverse magnetic fields are introduced. Compared to conventional imaginary-time evolution methods (ITEM), the 3D ResNet drastically reduces computational costs, offering a rapid and scalable solution for complex 3D multi-parameter nonlinear systems.

关键词: 3D ResNet, ground state phases prediction, data augmentation

Abstract: Machine learning has rapidly become a powerful tool for addressing challenges in ultracold atomic systems; however, its application to intricate three-dimensional (3D) systems remains relatively underexplored. In this study, we introduce a 3D residual network (3D ResNet) framework based on 3D convolutional neural networks (3D CNN) to predict ground states phases in 3D dipolar spinor Bose-Einstein condensates (BECs). Our results show that the 3D ResNet framework predicts ground states with high accuracy and efficiency across a broad parameter space. To enhance phase transition predictions, we incorporate data augmentation techniques, leading to a notable improvement in the model's performance. The method is further validated in more complex scenarios, particularly when transverse magnetic fields are introduced. Compared to conventional imaginary-time evolution methods (ITEM), the 3D ResNet drastically reduces computational costs, offering a rapid and scalable solution for complex 3D multi-parameter nonlinear systems.

Key words: 3D ResNet, ground state phases prediction, data augmentation

中图分类号: (Multicomponent condensates; spinor condensates)

67.85.Fg

03.75.Mn (Multicomponent condensates; spinor condensates) 05.30.Rt (Quantum phase transitions) 07.05.Mh (Neural networks, fuzzy logic, artificial intelligence)

Chengji Liao(廖承继), Tiantian Li(李甜甜), Xiao-Dong Bai(柏小东), and Yunbo Zhang(张云波). Three-dimensional ResNet for efficient prediction of ground state phases in multicomponent dipolar spinor BECs[J]. 中国物理B, 2025, 34(7): 76701-076701.

参考文献

[1] Santos L, Shlyapnikov G V, Zoller P, et al. 2002 Phys. Rev. Lett. 88 139904
[2] Lahaye T, Menotti C, Santos L, et al. 2009 Rep. Prog. Phys. 72 126401
[3] Tang Y, Kao W, Li K Y, et al. 2018 Phys. Rev. Lett. 120 230401
[4] Chomaz L, Baier S, Petter D, et al. 2016 Phys. Rev. X 6 041039
[5] Tanzi L, Roccuzzo S M, Lucioni E, et al. 2019 Nat. Phys. 574 382
[6] Baillie D and Blakie P B 2018 Phys. Rev. Lett. 121 195301
[7] Wenzel M, Böttcher F, Schmidt J N, et al. 2018 Phys. Rev. Lett. 121 030401
[8] Zhang F and Yin L 2023 Chin. Phys. Lett. 40 066701
[9] Zhang G, Lv L Z, Gao P, et al. 2023 Chin. Phys. B 32 110303
[10] Lan Z Z 2024 Chin. Phys. Lett. 41 090501
[11] Prasad S B, Bland T, Mulkerin B C, et al. 2019 Phys. Rev. A 100 023625
[12] Kumar R K, Tomio L, Malomed B A, et al. 2017 Phys. Rev. A 96 063624
[13] Ancilotto F and Reatto L 2024 Phys. Rev. A 110 013302
[14] Jiang Y, Qi R, Shi Z Y, et al. 2017 Phys. Rev. Lett. 118 080403
[15] Vengalattore M, Leslie S R, Guzman J, et al. 2008 Phys. Rev. Lett. 100 170403
[16] Li G, Zhao Z, Jiang X, et al. 2024 Phys. Rev. Lett. 133 053804
[17] Li S and Saito H 2024 Phys. Rev. Res. 6 L042049
[18] Yoǧurt T A, Tanyeri U, Keleş A, et al. 2003 Phys. Rev. A 108 033315
[19] Bigagli N, Yuan W, Zhang S, et al. 2024 Nature 631 289
[20] Yi S and Pu H 2006 Phys. Rev. Lett. 97 020401
[21] Kawaguchi Y, Saito H and Ueda M 2006 Phys. Rev. Lett. 97 130404
[22] Li T, Yi S and Zhang Y 2015 Phys. Rev. A 92 063603
[23] Chiofalo M L, Succi S and Tosi M P 2000 Phys. Rev. E 62 7438
[24] Motta M, Sun C, Tan A T K, et al. 2020 Nat. Phys. 16 205
[25] Silver D, Huang A, Maddison C, et al. 2016 Nature 529 484
[26] Krizhevsky A, Sutskever I and Hinton G E 2017 Commun. ACM 60 84
[27] Simonyan K and Zisserman A 2014 arXiv: 1409.1556
[28] Dosovitskiy A, Beyer L, Kolesnikov A, et al. 2020 arXiv:2010.11929
[29] Raissi M, Perdikaris P and Karniadakis G E 2019 J. Comput. Phys. 378 686
[30] Bar-Sinai Y, Hoyer S, Hickey J, et al. 2019 Proc. Natl. Acad. Sci. USA 116 15344
[31] Han J, Zhang L and Weinan E 2019 J. Comput. Phys. 399 108929
[32] Brunton S L and Kutz J N 2024 Nat. Comput. Sci. 4 483
[33] Vendeiro Z, Ramette J, Rudelis A, et al. 2022 Phys. Rev. Res. 4 043216
[34] Link M, Gao K, Kell A, et al. 2023 Phys. Rev. Lett. 130 203401
[35] Bai X D and Zhang D 2022 Phys. Rev. E 106 025305
[36] Guo S, Koh S M, Fritsch A R, et al. 2022 Phys. Rev. Res. 4 023163
[37] Bai X D and Zhang D 2021 Phys. Rev. A 104 063316
[38] Miao W, Linghu Z, Du Q, et al. 2023 Chin. Phys. Lett. 40 014301
[39] Raissi M, Perdikaris P and Karniadakis G E 2019 J. Comput. Phys. 378 686
[40] Wang N, Zhang D, Chang H, et al. 2020 J. Hydrol. 584 124700
[41] Karniadakis G E, Kevrekidis I G, Lu L, et al. 2021 Nat. Rev. Phys. 3 422
[42] Meng Z, Qian Q, Xu M, et al. 2023 Comput. Methods Appl. Mech. Eng. 414 116172
[43] Kuang J, Bai X D, Du W, et al. 2024 Results Phys. 64 107935
[44] Tian S, Cao C and Li B 2023 Results Phys. 52 106842
[45] Liu J, Wang W, Xia H, et al. 2024 Results Phys. 61 107716
[46] Ji S, XuW, Yang M, et al. 2013 IEEE Trans. Pattern Anal. Mach. Intell. 35 221
[47] Tran D, Bourdev L, Fergus R, et al. 2015 IEEE Int. Conf. Comput. Vis. 4489-4497
[48] He K, Zhang X, Ren S, et al. 2016 IEEE Conf. Comput. Vis. Pattern Recognit. 770-778
[49] Ho T L 1998 Phys. Rev. Lett. 81 742
[50] Ohmi T and Machida K 1998 J. Phys. Soc. Jpn. 67 1822
[51] Zhang J N, He L, Pu H, et al. 2009 Phys. Rev. A 79 033615
[52] Ç içek O, Abdulkadir A, Lienkamp S S, et al. 2016 MICCAI 9901
[53] Carreira J and Zisserman A 2017 IEEE Conf. Comput. Vis. Pattern Recognit. 4724-4733
[54] Li S, Song W, Fang L, et al. 2019 IEEE Trans. Geosci. Remote Sens. 57 6690
[55] Tran D, Bourdev L, Fergus R, et al. 2015 IEEE Int. Conf. Comput. Vis. 4489-4497
[56] Li Y, Zhang H and Shen Q 2017 Remote Sens. 9 67
[57] Peida W, Cui Z, Gan Z, et al. 2020 Sensors 20 1652
[58] Shorten C and Khoshgoftaar T M 2019 J. Big Data 6 60
[59] Bishop C M 1995 Neural Comput. 7 108

Three-dimensional ResNet for efficient prediction of ground state phases in multicomponent dipolar spinor BECs

Three-dimensional ResNet for efficient prediction of ground state phases in multicomponent dipolar spinor BECs

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

Metrics

本文评价

推荐阅读 0