PROGRESS OF CONVOLUTION NEURAL NETWORKS IN FLOW FIELD RECONSTRUCTION
-
摘要: 近年来, 随着深度学习在图像处理、语音识别、自动驾驶、自然语言处理等领域迅速发展, 该技术也被越来越广泛地应用于处理具有复杂非线性、高维度、大数据量等特点的流体力学方向. 传统的方法无法有效地处理这些庞大的数据, 深度学习因其具有强大的函数拟合能力, 可以从大量的数据中挖掘有用的信息. 当前, 流体力学深度学习技术有了初步的一些研究成果, 在流动信息特征提取、多源数据信息融合及流场的智能重构等方面具有重要的工程价值, 其应用潜力逐渐得到证实. 如何利用地面风洞试验、数值模拟及飞行试验获取的数据进行深入挖掘, 快速智能感知及重构流场, 可为主动流动控制提供重要指导. 本文主要从深度学习不同类型的网络结构出发探讨了卷积神经网络在流场重构中的研究进展, 文章首先介绍卷积神经网络的一些基本概念以及基本网络结构, 之后简要介绍流场超分辨率重构网络、端到端的映射网络、长短期记忆网络(LSTM)的基本结构与理论, 并详细归纳出他们的改进形式在流场重构领域的一系列研究进展与成果, 最后对文章做出总结并探讨了流场重构深度学习技术所面临的挑战与展望.Abstract: In recent years, with the rapid development of deep learning in image processing, speech recognition, automatic driving, natural language processing, and other fields, this technology is also more and more widely used to process fluid mechanics direction with the characteristics of complex non-linearity, high latitude and a large amount of data. Traditional methods can not effectively deal with these huge data. Due to its powerful function fitting ability, deep learning can mine useful information from a large amount of data. At present, the deep learning technology of fluid mechanics has some preliminary research results, it has important engineering value in flow information feature extraction, multi-sensor data information fusion and intelligent reconstruction of flow field, and its potential of application has been gradually confirmed. How to use the data obtained from ground wind tunnel test, numerical simulation and flight test to carry out in-depth mining, fast intelligent perception and reconstruction of flow field can provide important guidance for active flow control. Starting from different types of network structures of deep learning, this paper discusses the research progress of convolutional neural network in flow field reconstruction. Firstly, In this paper, we introduce some basic concepts and basic network structure of convolutional neural network, and then we briefly introduce the basic structure and theory of flow field super-resolution reconstruction network, end-to-end mapping network and short-term memory network (LSTM), A series of research progress and achievements of their improved forms in the field of flow field reconstruction are summarized in detail. Finally, we summarize the article and discuss the challenges and prospects of deep learning technology of flow field reconstruction.
-
图 7 基于最显着特征的现有单图像超分辨率技术的分类
Figure 7. Classification of existing single image super-resolution technology based on the most salient features
图 8 几种超分辨率重建方法的对比
Figure 8. Comparison of several super-resolution reconstruction methods
图 9 Bicubic、SRCNN、MPSRC重建结果的对比
Figure 9. Comparison of reconstruction results of Bicubic, SRCNN and MPSRC
图 13 模型重建和实验结果之间瞬时流场的比较: a) 楔角为0度, b)楔角为14度, c)楔角为20度
Figure 13. Comparison of instantaneous flow field between model reconstruction and experimental results: a) wedge angle of 0 degrees, b) wedge angle of 14 degrees, c) wedge angle of 20 degrees
图 14 速度场预测模型的CNN网络架构
Figure 14. CNN network architecture of velocity field prediction model
图 15 超声速燃烧室流场重构MBFCNN模型架构
Figure 15. MBFCNN model architecture for flow field reconstruction in supersonic combustor
16 不同雷诺数的CFD结果(左)与模型预测值(右)之间的瞬时流量比较
16. Comparison of instantaneous flow between CFD results of different Reynolds numbers (left) and model predicted values (right)
16 不同雷诺数的CFD结果(左)与模型预测值(右)之间的瞬时流量比较(续)
16. Comparison of instantaneous flow between CFD results of different Reynolds numbers (left) and model predicted values (right) (continued)
-
[1] 周铸, 黄江涛, 黄勇等. CFD技术在航空工程领域的应用, 挑战与发展. 航空学报, 2017, 03(3): 6-30 (Zhou Zhu, Huang Jiangtao, Huang Yong, et al. Application, challenge and development of CFD technology in aviation engineering. Acta Aeronautica et Astronautica Sinica, 2017, 03(3): 6-30 (in Chinese) [2] Li JW, Cheng YM, Chen KZ, et al. Cubic spline interpolation method of aerodynamic parameters based on AIWCPSO algorithm. Control and Decision, 2014, 29(1): 129-134 [3] Yuan C, Yang QZ, Chen LH. Flight simulation, and curve fitting for the characteristics of aircraft. Computer Simulation, 2008, 25(2): 53-56 [4] Raissi M, Wang ZC, Triantafyllou MS, et al. Deep learning of vortex-induced vibrations. Journal of Fluid Mechanics, 2019, 861: 119-137 doi: 10.1017/jfm.2018.872 [5] Huang J, Liu H, Cai W. Online in situ prediction of 3-D flame evolution from its history 2-D projections via deep learning. Journal of Fluid Mechanics, 2019, 875: R2 doi: 10.1017/jfm.2019.545 [6] Dowell EH. Eigenmode analysis in unsteady aerodynamics-reduced-order models. AIAA Journal, 1996, 34(8): 1578-1583 doi: 10.2514/3.13274 [7] Schmid PJ. Dynamic mode decomposition of numerical and experimental data. Journal of Fluid Mechanics, 2010, 656: 5-28 doi: 10.1017/S0022112010001217 [8] 罗杰, 段焰辉, 蔡晋生. 基于本征正交分解的流场快速预测方法研究. 航空工程进展, 2014, 5(3): 350-357 (Luo Jie, Duan Yanhui, Cai Jinsheng. Research on fast prediction method of flow field based on eigenorthogonal decomposition. Advances in Aeronautical Science and Engineering, 2014, 5(3): 350-357 (in Chinese) doi: 10.3969/j.issn.1674-8190.2014.03.014 [9] 侯强, 苏纬仪, 孙斐等. 高超声速乘波体扩容设计及流场快速预测. 航空动力学报, 2021, 36(3): 564-574 (Hou Qiang, Su Weiyi, Sun Fei, et al. Expansion design of hypersonic wave rider and rapid prediction of flow field. Journal of Aerospace Power, 2021, 36(3): 564-574 (in Chinese) [10] 李天一, Buzzicotti Michele, Biferale Luca等. Gappy POD方法重构湍流数据的研究. 力学学报, 2021, 53(10): 2703-2711 (Li Tianyi, Buzzicotti Michele, Biferale Luca, et al. Reconstruction of turbulent data with Gappy POD method. Chinese Journal of Theoretical and Applied Mechanics, 2021, 53(10): 2703-2711 (in Chinese) doi: 10.6052/0459-1879-21-464 [11] 张人会, 吴昊, 杨军虎等. 基于本征正交分解法的液环泵气液两相流场重构. 农业机械学报, 2017, 48(6): 6 (Zhang Renhui, Wu Hao, Yang Junhu, et al. Reconstruction of gas-liquid two-phase flow field of liquid ring pump based on eigen orthogonal decomposition method. Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(6): 6 (in Chinese) [12] Kutzj N. Deep learning in fluid dynamics. Journal of Fluid Mechanics, 2017, 814: 1-4 doi: 10.1017/jfm.2016.803 [13] 王建春, 晋国栋. 机器学习在力学模拟与控制中的应用专题序. 力学学报, 2021, 53(10): 2613-2615 (Wang Jianchun, Jin Guodong. Preface of theme articles on applications of machine learning to simulations and controls in mechanics. Chinese Journal of Theoretical and Applied Mechanics, 2021, 53(10): 2613-2615 (in Chinese) doi: 10.6052/0459-1879-21-501 [14] Guo X, Li W, Iorio F. Convolutional neural networks for steady flow approximation. Acm Sigkdd International Conference. 2016. p. 481–490 [15] Ling J, Kurzawski A, Templeton J. Reynolds averaged turbulence modelling using deep neural networks with embedded invariance. Journal of Fluid Mechanics. 2016, 807: 155–166 [16] Lecun Y, Bottou L. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 1998, 86(11): 2278-2324 doi: 10.1109/5.726791 [17] Krizhevsky A, Sutskever I, Hinton G. ImageNet Classification with Deep Convolutional Neural Networks. Advances in Neural Information Processing Systems, 2012, 25(2) [18] Szegedy C, Liu W, Jia Y, et al. Going Deeper with Convolutions. IEEE Computer Society, 2014: 1-9 [19] Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-Scale Image Recognition. Computer Science, 2014 [20] He K, Zhang X, Ren S, et al. Deep Residual Learning for Image Recognition. IEEE Computer Society, 2016: 770-778 [21] Woo S, Park J, Lee JY, et al. CBAM: Convolutional Block Attention Module. European Conference on Computer Vision. 2018: 3-19 [22] Fei W, Jiang M, Chen Q, et al. Residual attention network for image classification. IEEE, 2017: 6450-6458 [23] Chao D, Chen CL, He K, et al. Learning a deep convolutional network for image super-resolution. European Conference on Computer Vision. Springer, Cham, 2014: 184-199 [24] Zhou L, Lu X, Yang L. A local structure adaptive super-resolution reconstruction method based on BTV regularization. Multimedia Tools and Applications, 2014, 71(3): 1879-1892 doi: 10.1007/s11042-012-1311-x [25] Anwar S, Khan S, Barnes N. A deep journey into super-resolution: A survey. ACM Computing Surveys (CSUR) , 2020, 53(3): 1-34 [26] Zhang K, Zuo W, Chen Y, et al. Beyond a gaussian denoiser: Residual learning of deep CNN for image denoising. IEEE Transactions on Image Processing, 2017, 26(7): 3142-3155 doi: 10.1109/TIP.2017.2662206 [27] Lim B, Son S, Kim H, et al. Enhanced deep residual networks for single image super-resolution. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2017: 136-144 [28] Tai Y, Yang J, Liu X. Image super-resolution via deep recursive residual network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017: 3147-3155 [29] Lai WS, Huang JB, Ahuja N, et al. Deep laplacian pyramid networks for fast and accurate super-resolution. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017: 624-632 [30] Zhang Y, Tian Y, Kong Y, et al. Residual dense network for image super-resolution. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018: 2472-2481 [31] Shi W, Caballero J, Huszár F, et al. Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016: 1874-1883 [32] Choi JS, Kim M. A deep convolutional neural network with selection units for super-resolution. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2017: 1150-1156 [33] Anwar S, Barnes N. Densely residual laplacian super-resolution. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 44(3): 1192-1204 [34] Zhang K, Zuo W, Zhang L. Learning a single convolutional super-resolution network for multiple degradations. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018: 3262-3271 [35] Zhang Y, Li K, Li K, et al. Image super-resolution using very deep residual channel attention networks. Proceedings of the European Conference on Computer Vision (ECCV). 2018: 286-301 [36] Zhang K, Zuo W, Zhang L. Learning a single convolutional super-resolution network for multiple degradations. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018: 3262-3271 [37] Dong C, Loy CC, He K, et al. Image super-resolution using deep convolutional networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 38(2): 295-307 [38] Zhang K, Zuo W, Gu S, et al. Learning deep CNN denoiser prior for image restoration. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017: 3929-3938 [39] Liu B, Tang J, Huang H, et al. Deep learning methods for super-resolution reconstruction of turbulent flows. Physics of Fluids, 2020, 32(2): 025105 doi: 10.1063/1.5140772 [40] Kong C, Chang JT, Li YF, et al. Deep learning methods for super-resolution reconstruction of temperature fields in a supersonic combustor. AIP Advances, 2020, 10(11): 115021 doi: 10.1063/5.0030040 [41] Kai F, Fukagata K, Taira K. Super-resolution reconstruction of turbulent flows with machine learning. Journal of Fluid Mechanics, 2019, 870: 106-120 doi: 10.1017/jfm.2019.238 [42] Ferdian E, Suinesiaputra A, Dubowitz DJ, et al. 4 DFlowNet: Super-Resolution 4 D Flow MRI Using Deep Learning and Computational Fluid Dynamics. Frontiers in Physics, 2020, 8: 138 doi: 10.3389/fphy.2020.00138 [43] Kong C, Chang JT, Wang ZA, et al. Data-driven super-resolution reconstruction of supersonic flow field by convolutional neural networks. AIP Advances, 2021, 11(6): 065321 doi: 10.1063/5.0056569 [44] Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets. Advances in Neural Information Processing Systems, 2014: 27 [45] Xie Y, Franz E, Chu M, et al. tempogan: A temporally coherent, volumetric gan for super-resolution fluid flow. ACM Transactions on Graphics (TOG) , 2018, 37(4): 1-15 [46] Xu W, Luo W, Wang Y, et al. Data-driven three-dimensional super-resolution imaging of a turbulent jet flame using a generative adversarial network. Applied Optics, 2020, 59(19): 5729-5736 doi: 10.1364/AO.392803 [47] Lee J, Lee S, You D. Deep learning approach in multi-scale prediction of turbulent mixing-layer. Computer Vision and Pattern Recognition, 2018 [48] Liu Y, Wang Y, Deng L, et al. A novel in situ compression method for CFD data based on generative adversarial network. Journal of Visualization, 2019, 22(1): 95-108 doi: 10.1007/s12650-018-0519-x [49] Deng Z, He C, Liu Y, et al. Super-resolution reconstruction of turbulent velocity fields using a generative adversarial network-based artificial intelligence framework. Physics of Fluids, 2019, 31(12): 125111 doi: 10.1063/1.5127031 [50] Wu JL, Kashinath K, Albert A, et al. Enforcing Statistical Constraints in Generative Adversarial Networks for Modeling Chaotic Dynamical Systems. Journal of Computational Physics, 2020, 406: 109209 doi: 10.1016/j.jcp.2019.109209 [51] Lee S, You D. Data-driven prediction of unsteady flow over a circular cylinder using deep learning. Journal of Fluid Mechanics, 2019, 879: 217-254 doi: 10.1017/jfm.2019.700 [52] Lin J, Lensink K, Haber E. Fluid Flow Mass Transport for Generative Networks. Computer Vision and Pattern Recognition, 2019 [53] Kong C, Chang JT, Li YF, et al. Flowfield Reconstruction and Shock Train Leading Edge Detection in Scramjet Isolators. AIAA Journal, 2020, 58(9): 4068-4080 doi: 10.2514/1.J059302 [54] Kong C, Chang JT, Li YF, et al. A deep learning approach for the velocity field prediction in a scramjet isolator. Physics of Fluids, 2021, 33(2): 026103 doi: 10.1063/5.0039537 [55] Chen H, Guo MM, Tian Y, et al. Intelligent reconstruction of the flow field in a supersonic combustor based on deep learning. Physics of Fluids, 2022, 34(3): 035128 doi: 10.1063/5.0087247 [56] Marai GE, Monfort M, Luciani T, et al. A deep learning approach to identifying shock locations in turbulent combustion tensor fields. Modeling, Analysis, and Visualization of Anisotropy. Springer, Cham, 2017: 375-392 [57] Wang YQ, Wu Y, Shan L, et al. Flow field reconstruction method based on array neural network. Aeronautical Journal New Series, 2020, 125(1283): 1-21 [58] Hadikhani P, Borhani N, H Hashemi SM, et al. Learning from droplet flows in microfluidic channels using deep neural networks. Scientific Reports, 2019, 9(1): 1-7 [59] Bhatnagar S, Afshar Y, Pan S, et al. Prediction of aerodynamic flow fields using convolutional neural networks. Computational Mechanics, 2019, 64(2): 525-545 doi: 10.1007/s00466-019-01740-0 [60] Jin X, Cheng P, Chen WL. Prediction model of velocity field around circular cylinder over various Reynolds numbers by fusion convolutional neural networks based on pressure on the cylinder. Physics of Fluids, 2018, 30(4): 047105 doi: 10.1063/1.5024595 [61] 金晓威, 赖马树金, 李惠. 物理增强的流场深度学习建模与模拟方法. 力学学报, 2021, 53(10): 1-14 (Jin Xiaowei, Laima Shujin, Li Hui. Physics-enhanced deep learning methods for modelling and simulating flow fields. Chinese Journal of Theoretical and Applied Mechanics, 2021, 53(10): 1-14 (in Chinese) doi: 10.6052/0459-1879-21-222 [62] Hui X, Bai J, Wang H, et al. Fast pressure distribution prediction of airfoils using deep learning. Aerospace Science and Technology, 2020, 105: 105949 doi: 10.1016/j.ast.2020.105949 [63] Sekar V, Jiang Q, Shu C, et al. Fast flow field prediction over airfoils using deep learning approach. Physics of Fluids, 2019, 31(5): 057103 doi: 10.1063/1.5094943 [64] Umetani N, Bickel B. Learning three dimensional flow for interactive aerodynamic design. ACM Transactions on Graphics, 2018, 37(4): 1-10 [65] Umetani N. Exploring generative 3 D shapes using autoencoder networks. SIGGRAPH Asia 2017 Technical Briefs, 2017: 1-4 [66] Zhang Y, Sung WJ, Mavris DN. Application of convolutional neural network to predict airfoil lift coefficient. AIAA/ ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, 2018: 1903 [67] 王怡星, 李东风, 陈刚. 一种基于流场特征的气动力深度学习降阶模型. 第四届全国非定常空气动力学学术会议论文集, 2018: 135 (Wang Yixing, Li Dongfeng, Chen Gang. A deep learning reduced order model of aerodynamics based on flow field characteristics. Proceedings of the 4 th National Conference on Unsteady Aerodynamics, 2018: 135 (in Chinese) [68] 何磊, 钱炜祺, 汪清等. 机器学习方法在气动特性建模中的应用. 空气动力学学报, 2019, 37(3): 470-479 (He Lei, Qian weiqi, Wang Qing, et al. Application of machine learning method in aerodynamic characteristic modeling. Acta Aerodynamica Sinica, 2019, 37(3): 470-479 (in Chinese) [69] 杨从新, 凌祖光, 王岩等. 改进深度学习算法的对称翼型流场再现. 西安交通大学学报, 2021, 55(3): 20-28 (Yang Congxin, Ling Zuguang, Wang Yan, et al. Reconstruction of symmetrical airfoil flow field based on improved depth learning algorithm. Journal of Xi'an Jiaotong University, 2021, 55(3): 20-28 (in Chinese) [70] 惠心雨, 袁泽龙, 白俊强等. 基于深度学习的非定常周期性流动预测方法. 空气动力学学报, 2019, 37(03): 462-469 (Hui Xinyu, Yuan Zelong, Bai Junqiang, et al. Prediction method of unsteady periodic flow based on deep learning. Acta Aerodynamica Sinica, 2019, 37(03): 462-469 (in Chinese) [71] Zhao X, Du L, Peng X, et al. Research on refined reconstruction method of airfoil pressure based on compressed sensing. Theoretical and Applied Mechanics Letters, 2021, 11(2): 100223 doi: 10.1016/j.taml.2021.100223 [72] Hu JW, Zhang WW. Mesh-Conv: Convolution operator with mesh resolution independence for flow field modeling. Journal of Computational Physics, 2022, 452: 110896 doi: 10.1016/j.jcp.2021.110896 [73] Ribeiro MD, Rehman A, Ahmed S, et al. Deep CFD: efficient steady-state laminar flow approximation with deep convolutional neural networks. arXiv, 2020 [74] 尧少波, 何伟峰, 陈丽华等. 融合物理的神经网络方法在流场重建中的应用. 空气动力学学报, 2021, 39(s): 1-9 (Yao Shaobo, He Weifeng, Chen Lihua, et al. Application of neural network method integrating physics in flow field reconstruction. Acta Aerodynamica Sinica, 2021, 39(s): 1-9 (in Chinese) doi: 10.7638/kqdlxxb-2021.0332 [75] Raissi M, Yazdani A, Karniadakis GE. Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations. Science, 2020, 367(6481): 1026-1030 doi: 10.1126/science.aaw4741 [76] Lui HFS, Wolf WR. Construction of Reduced Order Models for Fluid Flows Using Deep Feedforward Neural Networks. Journal of Fluid Mechanics, 2019, 872: 963-994 doi: 10.1017/jfm.2019.358 [77] Hasegawa K, Fukami K, Murata T, et al. Machine-learning-based reduced-order modeling for unsteady flows around bluff bodies of various shapes. Theoretical and Computational Fluid Dynamics, 2020, 34(4): 367-383 doi: 10.1007/s00162-020-00528-w [78] Hu L, Wang W, Xiang Y, et al. Flow field reconstructions with gans based on radial basis functions. IEEE Transactions on Aerospace and Electronic Systems, 2022 [79] Ma W, Zhang J, Yu J. Non-intrusive reduced order modeling for flowfield reconstruction based on residual neural network. Acta Astronautica, 2021, 183: 346-362 doi: 10.1016/j.actaastro.2020.11.050 [80] Brahmachary S, Bhagyarajan A, Ogawa H, Hideaki Ogawa. Fast estimation of internal flowfields in scramjet intakes via reduced-order modeling and machine learning. Physics of Fluids, 2021, 33(10): 106110 doi: 10.1063/5.0064724 [81] Hochreiter S, Schmidhuber J. Long Short-Term Memory. Neural Computation, 1997, 9(8): 1735-1780 doi: 10.1162/neco.1997.9.8.1735 [82] Mohan AT, Gaitonde DV. A deep learning based approach to reduced order modeling for turbulent flow control using LSTM neural networks. arXiv, 2018 [83] Pawar S, Rahman SM, Vaddireddy H, et al. A deep learning enabler for nonintrusive reduced order modeling of fluid flows. Physics of Fluids, 2019, 31(8): 085101 doi: 10.1063/1.5113494 [84] Deng Z, Chen Y, Liu Y, et al. Time-resolved turbulent velocity field reconstruction using a long short-term memory (LSTM)-based artificial intelligence framework. Physics of Fluids, 2019, 31(7): 075108 doi: 10.1063/1.5111558 [85] Mohan A, Daniel D, Chertkov M, et al. Compressed convolutional LSTM: An efficient deep learning framework to model high fidelity 3 D turbulence. arXiv, 2019 [86] Li YF, Chang JT, Wang ZA, et al. An efficient deep learning framework to reconstruct the flow field sequences of the supersonic cascade channel. Physics of Fluids, 2021, 33(5): 056106 doi: 10.1063/5.0048170 [87] Li YF, Wang ZA, Jiang W, et al. Research on time sequence prediction of the flow field structure of supersonic cascade channels in wide range based on artificial neural network. Physics of Fluids, 2022, 34(1): 016106 doi: 10.1063/5.0073109 [88] Han R, Wang Y, Zhang Y, et al. A novel spatial temporal prediction method for unsteady wake flows based on hybrid deep neural network. Physics of Fluids, 2019, 31(12): 127101 doi: 10.1063/1.5127247 [89] 尹云飞, 林跃江, 黄发良等. 基于趋势特征向量的火灾烟气流动与温度分布预测. 计算机科学, 2021, 48(7): 299-307 (Yin Yunfei, Lin Yuejiang, Huang Faliang, et al. Prediction of fire smoke flow and temperature distribution based on trend eigenvector. Computer Science, 2021, 48(7): 299-307 (in Chinese) [90] Sabour S, Frosst N, Hinton GE. Dynamic routing between capsules, Advances in Neural Information Processing Systems, 2017, 30 [91] Vaswani A, Shazeer N, Parmar N, et al. Attention Is All You Need. Advances in Neural Information Processing Systems, 2017: 30 [92] Alaparthi S, Mishra M. Bidirectional Encoder Representations from Transformers (BERT): A sentiment analysis odyssey. arXiv, 2020 [93] Zhou J, Cui G, Hu S, et al. Graph neural networks: A review of methods and applications. AI Open, 2020, 1: 57-81 doi: 10.1016/j.aiopen.2021.01.001 -
下载:
点击查看大图
图(19)
计量
- 文章访问数: 29
- HTML全文浏览量: 8
- PDF下载量: 4
- 被引次数: 0
