期刊：
Machine Learning: Science and Technology,2024年5(1):015033 ISSN：2632-2153
通讯作者：
Li, W
作者机构：
[Liu, Feiyi; Li, W; Wang, Yanyang; Li, Wei; Shen, Jianmin] Cent China Normal Univ, Key Lab Quark & Lepton Phys MOE, Wuhan 430079, Peoples R China.;[Liu, Feiyi; Li, W; Wang, Yanyang; Li, Wei; Shen, Jianmin] Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.;[Liu, Feiyi] Eotvos Lorand Univ, Inst Phys, 1-A Pazmany P Setany, H-1117 Budapest, Hungary.;[Shen, Jianmin] Baoshan Univ, Coll Engn & Technol, Baoshan, Peoples R China.
通讯机构：
[Li, W ] C;Cent China Normal Univ, Key Lab Quark & Lepton Phys MOE, Wuhan 430079, Peoples R China.;Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.
关键词：
Convolutional neural networks;Monte Carlo methods;Random processes;Solvents;Supervised learning;Unsupervised learning;Auto encoders;Branching annihilating random walk;Convolutional neural network;Directed percolation;Equilibrium systems;Machine-learning;Nonequilibrium phase transitions;Nonequilibrium system;Random Walk;Supervised and unsupervised learning;Convolution
摘要：
<jats:title>Abstract</jats:title>
<jats:p>Machine learning (ML) of phase transitions (PTs) has gradually become an effective approach that enables us to explore the nature of various PTs more promptly in equilibrium and nonequilibrium systems. Unlike equilibrium systems, non-equilibrium systems display more complicated and diverse features because of the extra dimension of time, which is not readily tractable, both theoretically and numerically. The combination of ML and most renowned nonequilibrium model, directed percolation (DP), led to some significant findings. In this study, ML is applied to <jats:inline-formula>
<jats:tex-math><?CDATA $(1+1)$?></jats:tex-math>
<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll">
<mml:mo stretchy="false">(</mml:mo>
<mml:mn>1</mml:mn>
<mml:mo>+</mml:mo>
<mml:mn>1</mml:mn>
<mml:mo stretchy="false">)</mml:mo>
</mml:math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="mlstad27e2ieqn2.gif" xlink:type="simple" />
</jats:inline-formula>-d, even offspring branching annihilating random walks (BAW), whose universality class is not DP-like. The supervised learning of <jats:inline-formula>
<jats:tex-math><?CDATA $(1+1)$?></jats:tex-math>
<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll">
<mml:mo stretchy="false">(</mml:mo>
<mml:mn>1</mml:mn>
<mml:mo>+</mml:mo>
<mml:mn>1</mml:mn>
<mml:mo stretchy="false">)</mml:mo>
</mml:math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="mlstad27e2ieqn3.gif" xlink:type="simple" />
</jats:inline-formula>-d BAW via convolutional neural networks (CNN) results in a more accurate prediction of the critical point than the Monte Carlo (MC) simulation for the same system sizes. The dynamic exponent <jats:italic>z</jats:italic> and spatial correlation length correlation exponent <jats:inline-formula>
<jats:tex-math><?CDATA $\nu_{\perp}$?></jats:tex-math>
<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll">
<mml:msub>
<mml:mi>ν</mml:mi>
<mml:mrow>
<mml:mo>⊥</mml:mo>
</mml:mrow>
</mml:msub>
</mml:math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="mlstad27e2ieqn4.gif" xlink:type="simple" />
</jats:inline-formula> were also measured and found to be consistent with their respective theoretical values. Furthermore, the unsupervised learning of <jats:inline-formula>
<jats:tex-math><?CDATA $(1+1)$?></jats:tex-math>
<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll">
<mml:mo stretchy="false">(</mml:mo>
<mml:mn>1</mml:mn>
<mml:mo>+</mml:mo>
<mml:mn>1</mml:mn>
<mml:mo stretchy="false">)</mml:mo>
</mml:math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="mlstad27e2ieqn5.gif" xlink:type="simple" />
</jats:inline-formula>-d BAW via an autoencoder (AE) gives rise to a transition point, which is the same as the critical point. The latent layer of AE, through a single neuron, can be regarded as the order parameter of the system being properly re-scaled. Therefore, we believe that ML has exciting application prospects in reaction-diffusion systems such as BAW and DP.</jats:p>
作者机构：
[Liu, Zhihang; Li, W; Li, Wei; Yang, Yuxiang] Cent China Normal Univ, Key Lab Quark & Lepton Phys, MOE, Wuhan 430079, Peoples R China.;[Liu, Zhihang; Li, W; Li, Wei; Yang, Yuxiang] Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.
通讯机构：
[Li, W ] C;Cent China Normal Univ, Key Lab Quark & Lepton Phys, MOE, Wuhan 430079, Peoples R China.;Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.
摘要：
The rich-club phenomenon, which provides information about the association between nodes, is a useful method to study the hierarchy structure of networks. In this work, we explore the behavior of rich-club coefficient (RCC) in scale-free networks, and find that the degree-based RCC is a power function of degree centrality with power exponent beta and the betweenness-based RCC is a linear function of betweenness centrality with slope theta. Moreover, we calculate the value of RCC in a BA network by deleting nodes and obtain a general expression for RCC as a function of node sequence. On this basis, the solution of RCC for centrality is also obtained, which shows that the curve of RCC is determined by the centrality distribution. In the numerical simulation, we observe: beta and gamma (the degree distribution exponent) increase together, theta increases with the average degree and decreases to convergence as gamma increases.
期刊：
Physica A-Statistical Mechanics and its Applications,2024年637:129533 ISSN：0378-4371
通讯作者：
Liu, FY
作者机构：
[Yang, Chunbin; Liu, Feiyi; Deng, Weibing; Li, Wei; Chen, Xiangna; Liu, FY] Cent China Normal Univ, Key Lab Quark & Lepton Phys MOE, Wuhan 430079, Peoples R China.;[Liu, Feiyi; Deng, Weibing; Li, Wei; Liu, FY] Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.;[Liu, Feiyi; Papp, Gabor; Liu, FY] Eotv Lorand Univ, Inst Phys, 1-A Pazmany P Setany, H-1117 Budapest, Hungary.;[Chen, Shiyang] Baoshan Univ, Sch Engn & Technol, Baoshan 678000, Peoples R China.;[Shen, Jianmin] Max Planck Inst Math Sci, D-04103 Leipzig, Germany.
通讯机构：
[Liu, FY ] C;Cent China Normal Univ, Key Lab Quark & Lepton Phys MOE, Wuhan 430079, Peoples R China.;Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.;Eotv Lorand Univ, Inst Phys, 1-A Pazmany P Setany, H-1117 Budapest, Hungary.
摘要：
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.
作者机构：
[Li, W; Li, Wei; Tuo, Kui] Cent China Normal Univ, Key Lab Quark & Lepton Phys, MOE, Wuhan 430079, Peoples R China.;[Li, W; Li, Wei; Tuo, Kui] Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.;[Deng, Shengfeng] Shaanxi Normal Univ, Sch Phys & Informat Technol, Xian 710061, Peoples R China.;[Zhu, Yueying] Wuhan Text Univ, Res Ctr Appl Math & Interdisciplinary Sci, Wuhan 430073, Peoples R China.
通讯机构：
[Li, W ] C;Cent China Normal Univ, Key Lab Quark & Lepton Phys, MOE, Wuhan 430079, Peoples R China.;Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.
摘要：
The Domany-Kinzel (DK) model encompasses several types of nonequilibrium phase transitions, depending on the selected parameters. We apply supervised, semisupervised, and unsupervised learning methods to studying the phase transitions and critical behaviors of the (1 + 1)-dimensional DK model. The supervised and the semisupervised learning methods permit the estimations of the critical points, the spatial and temporal correlation exponents, concerning labeled and unlabeled DK configurations, respectively. Furthermore, we also predict the critical points by employing principal component analysis and autoencoder. The PCA and autoencoder can produce results in good agreement with simulated stationary particle number density.
期刊：
Physica A-Statistical Mechanics and its Applications,2023年617:128666 ISSN：0378-4371
通讯作者：
Liu, FY
作者机构：
[Yang, Chunbin; Liu, Feiyi; Chen, Shiyang; Deng, Weibing; Li, Wei; Chen, Xiangna; Liu, FY] Cent China Normal Univ, Key Lab Quark & Lepton Phys MOE, Wuhan 430079, Peoples R China.;[Yang, Chunbin; Liu, Feiyi; Chen, Shiyang; Deng, Weibing; Li, Wei; Chen, Xiangna; Liu, FY] Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.;[Liu, Feiyi; Papp, Gabor; Liu, FY] Eotvos Lorand Univ, Inst Phys, 1-A Pazmany P Setany, H-1117 Budapest, Hungary.;[Li, Wei] Max Planck Inst Math Sci, D-04103 Leipzig, Germany.;[Shen, Jianmin] Baoshan Univ, Sch Engn & Technol, Baoshan 678000, Peoples R China.
通讯机构：
[Liu, FY ] C;Cent China Normal Univ, Key Lab Quark & Lepton Phys MOE, Wuhan 430079, Peoples R China.;Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.;Eotvos Lorand Univ, Inst Phys, 1-A Pazmany P Setany, H-1117 Budapest, Hungary.
摘要：
A transfer learning method, Domain Adversarial Neural Network (DANN), is introduced to study the phase transition of two-dimensional q-state Potts model. With the DANN, we only need to choose a few labeled configurations automatically as input data, then the critical points can be obtained after training the algorithm. By an additional iterative process, the critical points can be captured to comparable accuracy to Monte Carlo simulations as we demonstrate it for q = 3,4, 5, 7 and 10. The type of phase transition (first or second-order) is also determined at the same time. Meanwhile, for the second -order phase transition at q = 3, we can calculate the critical exponent & nu; by data collapse. Furthermore, compared to the traditional supervised learning, we found the DANN to be more accurate with lower cost.& COPY; 2023 Elsevier B.V. All rights reserved.
期刊：
Physica A-Statistical Mechanics and its Applications,2023年609:128329 ISSN：0378-4371
通讯作者：
Yueying Zhu
作者机构：
[Zhu, Yueying; Jiang, Jian] Wuhan Text Univ, Res Ctr Nonlinear Sci, Sch Math & Phys Sci, Wuhan 430200, Peoples R China.;[Li, Wei] Cent China Normal Univ, Complex Sci Ctr, Wuhan 430079, Peoples R China.;[Li, Wei] Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.;[Li, Wei] Max Planck Inst Math Sci, Inselst 22, D-04103 Leipzig, Germany.
通讯机构：
[Yueying Zhu] R;Research Center of Nonlinear Science, School of Mathematical & Physical Sciences, Wuhan Textile University, 430200 Wuhan, China
摘要：
The Hegselmann-Krause (HK) model allows one to characterize the continuous change of agent opinions with the bounded confidence threshold epsilon. To consider the heterogeneity of agents in characteristics, we study the HK model on homogeneous and heterogeneous networks by introducing a kind of smart agent. Different from the averaging rule in opinion update of HK model, smart agents will consider, in updating their opinions, the environmental influence following the fact that the agent behavior is often coupled with environmental changes. The environment is characterized by a parameter that represents the biased resource allocation between different cliques. We focus on the critical behavior of the underlying system. A phase transition point separating a complete consensus from the coexistence of different opinions is identified, which occurs at a critical value epsilon c for the bounded confidence threshold. We state analytically that epsilon c can take only one of two possible values, depending on the behavior of the average degree ka of a social graph, when agents are homogeneous in characteristics. Results also suggest that the phase transition point weakly depends on the network structure but is strongly correlated with the fraction of smart agents and the environmental parameter. We finally establish the finite size scaling law that stresses the role that the system size has in the underlying opinion dynamics. Meanwhile, introducing smart agents does not change the functional dependence between the time to reach a complete consensus and the system size. However, it can drive a complete consensus to be reached faster, for homogeneous networks that are far from the mean field limit.(c) 2022 Elsevier B.V. All rights reserved.
作者机构：
[Fei Ma] Key Laboratory of Quark and Lepton Physics (MOE) and Institute of Particle Physics, Central China Normal University, WuHan, 430079, China;Institute for Physics, Eötvös Loránd University, 1/A Pázmány P. Sétány, H-1117, Budapest, Hungary;Max-Planck-Institute for Mathematics in the Sciences, 04103 Leipzig, Germany;[Feiyi Liu] Key Laboratory of Quark and Lepton Physics (MOE) and Institute of Particle Physics, Central China Normal University, WuHan, 430079, China<&wdkj&>Institute for Physics, Eötvös Loránd University, 1/A Pázmány P. Sétány, H-1117, Budapest, Hungary;[Wei Li] Key Laboratory of Quark and Lepton Physics (MOE) and Institute of Particle Physics, Central China Normal University, WuHan, 430079, China<&wdkj&>Max-Planck-Institute for Mathematics in the Sciences, 04103 Leipzig, Germany
摘要：
Recently, methods of graph neural networks (GNNs) have been applied to solving the problems in high-energy physics (HEP) and have shown its great potential for quark-gluon tagging with graph representation of jet events. In this paper, we introduce an approach of GNNs combined with a Haar pooling operation to analyze the events, called Haar pooling message passing neural network (HMPNet). In HMPNet, Haar pooling not only extracts the features of graph, but embeds additional information obtained by clustering of k means of different particle features. We construct Haar pooling from five different features: absolute energy logE , transverse momentum logpT , relative coordinates (Δη,Δϕ) , the mixed ones (logE,logpT) , and (logE,logpT,Δη,Δϕ) . The results show that an appropriate selection of information for Haar pooling enhances the accuracy of quark-gluon tagging, as adding extra information of logPT to the HMPNet outperforms all the others, whereas adding relative coordinates information (Δη,Δϕ) is not very effective. This implies that, by adding effective particle features from Haar pooling, one can achieve much better results than that which a solely pure message passing neutral network can do, which demonstrates significant improvement of feature extraction via the pooling process. Finally, we compare the HMPNet study, ordering by pT , with other studies and prove that the HMPNet is also a good choice of GNN algorithms for jet tagging.
通讯机构：
[Yiwen Tang] I;Institute of Nano-Science and Technology, College of Physical Science and Technology, Central China Normal University, Wuhan 430079, China
摘要：
Polyoxometalates (POM) have a wide range of applications in electrochemistry, catalysis, and energy storage due to their variable structure and nature. Here, we designed a novel 3D porous network of ultrathin V6O13- POM nanosheets on carbon cloth (V6O13-POM//CC) for the first time by one-pot hydrothermal method with the help of phosphotungstic acid. Serving as inorganic ligands, the anions of phosphotungstic acid co-ordinate with V cations, facilitating the assembly of nanoclusters and guiding to control of the morphology of V6O13-POM composites, which increases the specific surface area and multiple active sites for the re-action of lithium ions batteries (LIBs). Furthermore, the synergistic effect of V6O13 and POM improves the stability of the structure and increases the capacity of the electrode in LIBs. V6O13-POM//CC as an anode material exhibited a high reversible capacity of 2.03 mAh cm-2 at a high current density of 2 mA cm-2 after 120 cycles. To illustrate the generality of the synthesis strategy, we also obtained MoO3-POM//CC via a similar method. By using V6O13-POM//CC and MoO3-POM//CC as the anode materials, LIBs exhibited high reversible areal specific capacity and excellent cycling stability which are expected to achieve commercial applications.(c) 2022 Elsevier B.V. All rights reserved.
期刊：
Journal of Alloys and Compounds,2022年895:162535 ISSN：0925-8388
通讯作者：
Tang, Yiwen;Wang, Hai
作者机构：
[Chen, Mingyue; Tang, Yiwen; Qi, Pengcheng; Liu, Gaofu; Huang, Chuqiang; Lu, Yu; Li, Wenhui] Cent China Normal Univ, Inst Nanosci & Nanotechnol, Coll Phys Sci & Technol, Wuhan 430079, Peoples R China.;[Wang, Hai] China Univ Geosci, Sch Math & Phys, Wuhan 430074, Peoples R China.
通讯机构：
[Tang, Yiwen; Wang, Hai] C;Cent China Normal Univ, Inst Nanosci & Nanotechnol, Coll Phys Sci & Technol, Wuhan 430079, Peoples R China.;China Univ Geosci, Sch Math & Phys, Wuhan 430074, Peoples R China.
关键词：
Activated carbon;Ammonia;Cathodes;Cobalt compounds;Doping (additives);Nickel;Nickel compounds;Supercapacitor;Transition metal oxides;Electrode material;Higher energy density;Low Power;N-doped;Nanosheet arrays;Ni foam;NiCoO2 nanosheet;Positive electrodes;Rate performance;Transition-metal oxides;Nanosheets
摘要：
Transition metal oxides as the electrode material of supercapacitors have been widely studied due to their high energy density. However, relatively low power density resulting from poor conductivity of the metal oxides limits their application. In this paper, a new N-doped NiCoO2 nanosheet array on Ni foam was synthesized through an ammonia-induced reduction strategy. During the ventilation of NH3 in the ammonia annealing stage, N was doped into the structure of NiCoO2, which leads to increased concentration of oxygen vacancies and improved conductivity. As a positive electrode, N-doped NiCoO2 exhibits excellent rate performance (1449.3 F g(-1) at 1 A g(-1) and 1190.4 F g(-1) at 50 A g(-1)) and impressive cycle stability (the capacitance retention ratio is 92% after 5000 cycles at the current density of 40 A g(-1)) compared to undoped NiCoO2 nanosheet array electrode. Assembling it as the positive electrode and activated carbon as the negative electrode, the aqueous asymmetric supercapacitor exhibits high power density, energy density, and excellent rate performance. The remarkable energy storage performances make N-doped NiCoO2 nanosheet arrays material (N-NiCoO2) have a broad application prospect in the field of supercapacitors. (C) 2021 Elsevier B.V. All rights reserved.
通讯机构：
[Yue Hu] M;[Yiwen Tang] D;Michael Grätzel Center for Mesoscopic Solar Cells, Wuhan National Laboratory for Optoelectronics, Huazhong University of Science and Technology, Wuhan, 430074 China<&wdkj&>Department Nano-Science & Technology, College of Physics and Technology, Central China Normal University, Wuhan, 430079 China
关键词：
crystallization quality;CsPbBr3;energy band alignment;europium ions;perovskite solar cells
摘要：
The high‐valence Eu3+ was applied in carbon‐based printable mesoscopic inorganic CsPbBr3 perovskite solar cells (PSCs) for the first time. The high open‐circuit voltage and stability of the device demonstrate that the CsPbBr3 PSC is a promising device to drive the water electrolysis device. The all‐inorganic CsPbBr3 perovskite exhibits the possibility of overcoming the substantial nonideal thermal, humidity, and photostability of hybrid organic–inorganic perovskite solar cells (PSCs) in photoelectronic devices. Specifically, the rapid development of CsPbBr3 perovskite has delivered device efficiencies >10%. However, the mismatched energy band alignment and bad crystallization quality are still potential obstacles for the superior performance of PSCs. Herein, by employing n‐type doping, trivalent europium cation is successfully introduced into the CsPbBr3 lattice. The better energy‐level alignment leads to further reduction of voltage losses. Besides, the large and uniform grains resulting from the improvement of crystallization after doping decrease the grain boundaries and reduce the nonradiative recombination center. The quality of the film improves substantially, which significantly enhances the photoabsorption and the short‐circuit current density. The efficiency of the carbon‐based printable mesoscopic PSCs is improved from 7.5% to 8.06% with 3 mol% Eu3+ doping, resulting in high open‐circuit voltage of 1.41 V. Based on the device with effective area of 1 cm2 and 60.075 cm2, the record power conversion efficiency of 5.41% and 1.14% is obtained. The device also displays excellent stability with driving water electrolysis.
作者机构：
[Liu, Feiyi; Chen, Shiyang; Shen, Jianmin; Li, Wei; Xu, Dian] Cent China Normal Univ, Key Lab Quark & Lepton Phys MOE, Wuhan 430079, Peoples R China.;[Liu, Feiyi; Chen, Shiyang; Shen, Jianmin; Li, Wei; Xu, Dian] Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.;[Deng, Shengfeng] Ctr Energy Res, Inst Tech Phys & Mat Sci, H-1121 Budapest, Hungary.;[Liu, Feiyi] Eotvos Lorand Univ, Inst Phys, 1-A Pazmany P Setany, H-1117 Budapest, Hungary.
通讯机构：
[Li, Wei] K;Key Laboratory of Quark and Lepton Physics (MOE) and Institute of Particle Physics, Central China Normal University, Wuhan, 430079, China.
摘要：
<jats:title>Abstract</jats:title><jats:p>The pair-contact process with diffusion (PCPD), a generalized model of the ordinary pair-contact process (PCP) without diffusion, exhibits a continuous absorbing phase transition. Unlike the PCP, whose nature of phase transition is clearly classified into the directed percolation (DP) universality class, the model of PCPD has been controversially discussed since its infancy. To our best knowledge, there is so far no consensus on whether the phase transition of the PCPD falls into the unknown university classes or else conveys a new kind of non-equilibrium phase transition. In this paper, both unsupervised and supervised learning are employed to study the PCPD with scrutiny. Firstly, two unsupervised learning methods, principal component analysis (PCA) and autoencoder, are taken. Our results show that both methods can cluster the original configurations of the model and provide reasonable estimates of thresholds. Therefore, no matter whether the non-equilibrium lattice model is a random process of unitary (for instance the DP) or binary (for instance the PCP), or whether it contains the diffusion motion of particles, unsupervised learning can capture the essential, hidden information. Beyond that, supervised learning is also applied to learning the PCPD at different diffusion rates. We proposed a more accurate numerical method to determine the spatial correlation exponent <jats:inline-formula><jats:alternatives><jats:tex-math>$$\nu _{\perp }$$</jats:tex-math><mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML">
<mml:msub>
<mml:mi>ν</mml:mi>
<mml:mo>⊥</mml:mo>
</mml:msub>
</mml:math></jats:alternatives></jats:inline-formula>, which, to a large degree, avoids the uncertainty of data collapses through naked eyes.</jats:p>
作者机构：
[Yang, Chunbin; Liu, Feiyi; Chen, Shiyang; Shen, Jianmin; Li, Wei; Chen, Xiangna; Xu, Dian] Cent China Normal Univ, Key Lab Quark & Lepton Phys MOE, Wuhan 430079, Peoples R China.;[Yang, Chunbin; Liu, Feiyi; Chen, Shiyang; Shen, Jianmin; Li, Wei; Chen, Xiangna; Xu, Dian] Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.;[Liu, Feiyi; Papp, Gabor] Eotvos Lorand Univ, Inst Phys, 1-A Pazmany P Setany, H-1117 Budapest, Hungary.;[Deng, Shengfeng] Ctr Energy Res, Inst Tech Phys & Mat Sci, H-1121 Budapest, Hungary.;[Li, Wei] Max Planck Inst Math Sci, D-04103 Leipzig, Germany.
摘要：
The latest advances of statistical physics have shown remarkable performance of machine learning in identifying phase transitions. In this paper, we apply domain adversarial neural network (DANN) based on transfer learning to studying nonequilibrium and equilibrium phase transition models, which are percolation model and directed percolation (DP) model, respectively. With the DANN, only a small fraction of input configurations (two-dimensional images) needs to be labeled, which is automatically chosen, to capture the critical point. To learn the DP model, the method is refined by an iterative procedure in determining the critical point, which is a prerequisite for the data collapse in calculating the critical exponent ?????. We then apply the DANN to a two-dimensional site percolation with configurations filtered to include only the largest cluster which may contain the information related to the order parameter. The DANN learning of both models yields reliable results which are comparable to the ones from Monte Carlo simulations. Our study also shows that the DANN can achieve quite high accuracy at much lower cost, compared to the supervised learning.
作者：
Agakishiev, H.*;Aggarwal, M. M.;Ahammed, Z.;Alakhverdyants, A., V;Alekseev, I;...
期刊：
中国物理C,2021年45(4):198-241 ISSN：1674-1137
通讯作者：
Agakishiev, H.
作者机构：
[Kechechyan, A.; Agakishiev, H.; Averichev, G. S.; Efimov, L. G.; Alakhverdyants, A., V; Kizka, V; Rogachevskiy, O., V; Panebratsev, Y.; Shahaliev, E.; Fedorisin, J.; Bunzarov, I; Zoulkarneev, R.; Zoulkarneeva, Y.; Tokarev, M.; Filip, P.; Lednicky, R.; Dedovich, T. G.; Vokal, S.] Joint Inst Nucl Res, Dubna 141980, Russia.;[Underwood, D. G.; Krueger, K.; Bridgeman, A.; Spinka, H. M.] Argonne Natl Lab, Argonne, IL 60439 USA.;[Ullrich, T.; Pile, P.; Lamont, M. A. C.; Ljubicic, T.; Le Vine, M. J.; Xu, Z.; Ogawa, A.; Ruan, L.; Tang, A. H.; Videbaek, F.; Fisyak, Y.; Beavis, D. R.; Lauret, J.; Bland, L. C.; Longacre, R. S.; Webb, J. C.; Christie, W.; Arkhipkin, D.; Fine, V; Debbe, R. R.; Lebedev, A.; Burton, T. P.; Dunlop, J. C.; Gordon, A.; Landgraf, J. M.; Love, W. A.; Yip, K.; Didenko, L.; Guryn, W.; Van Buren, G.; Lee, J. H.] Brookhaven Natl Lab, Upton, NY 11973 USA.;[Perkins, C.; Crawford, H. J.; Engelage, J.; Judd, E. G.] Univ Calif Berkeley, Berkeley, CA 94720 USA.;[Sanchez, M. Calderon de la Barca; Draper, J. E.; Cebra, D.; Brovko, S. G.; Mall, O., I; Sangaline, E.; Reed, R.; Romero, J. L.; Salur, S.; Haag, B.; Liu, H.] Univ Calif Davis, Davis, CA 95616 USA.
通讯机构：
[Agakishiev, H.] J;Joint Inst Nucl Res, Dubna 141980, Russia.
关键词：
relativistic heavy ion collisions;dihadron correlations;jet-medium interactions;anisotropic flow background;event plane
摘要：
<jats:title>Abstract</jats:title>
<jats:p>Dihadron azimuthal correlations containing a high transverse momentum (
<jats:inline-formula>
<jats:tex-math><?CDATA $ p_{T} $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M2.jpg" xlink:type="simple" />
</jats:inline-formula>) trigger particle are sensitive to the properties of the nuclear medium created at RHIC through the strong interactions occurring between the traversing parton and the medium, i.e. jet-quenching. Previous measurements revealed a strong modification to dihadron azimuthal correlations in Au+Au collisions with respect to <jats:italic>p</jats:italic>+<jats:italic>p</jats:italic> and <jats:italic>d</jats:italic>+Au collisions. The modification increases with the collision centrality, suggesting a path-length or energy density dependence to the jet-quenching effect. This paper reports STAR measurements of dihadron azimuthal correlations in mid-central (20%-60%) Au+Au collisions at
<jats:inline-formula>
<jats:tex-math><?CDATA $ \sqrt{s_{\rm{NN}}} = 200 $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M3.jpg" xlink:type="simple" />
</jats:inline-formula> GeV as a function of the trigger particle's azimuthal angle relative to the event plane,
<jats:inline-formula>
<jats:tex-math><?CDATA $ \phi_{s} = | \phi_{t}- \psi_{{\rm{EP}}}| $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M4.jpg" xlink:type="simple" />
</jats:inline-formula>. The azimuthal correlation is studied as a function of both the trigger and associated particle
<jats:inline-formula>
<jats:tex-math><?CDATA $ p_{T} $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M5.jpg" xlink:type="simple" />
</jats:inline-formula>. The subtractions of the combinatorial background and anisotropic flow, assuming Zero Yield At Minimum (ZYAM), are described. The correlation results are first discussed with subtraction of the even harmonic (elliptic and quadrangular) flow backgrounds. The away-side correlation is strongly modified, and the modification varies with
<jats:inline-formula>
<jats:tex-math><?CDATA $ \phi_{s} $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M6.jpg" xlink:type="simple" />
</jats:inline-formula>, with a double-peak structure for out-of-plane trigger particles. The near-side ridge (long range pseudo-rapidity
<jats:inline-formula>
<jats:tex-math><?CDATA $ \Delta\eta $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M7.jpg" xlink:type="simple" />
</jats:inline-formula> correlation) appears to drop with increasing
<jats:inline-formula>
<jats:tex-math><?CDATA $ \phi_{s} $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M8.jpg" xlink:type="simple" />
</jats:inline-formula> while the jet-like component remains approximately constant. The correlation functions are further studied with the subtraction of odd harmonic triangular flow background arising from fluctuations. It is found that the triangular flow, while responsible for the majority of the amplitudes, is not sufficient to explain the
<jats:inline-formula>
<jats:tex-math><?CDATA $ \phi_{s} $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M9.jpg" xlink:type="simple" />
</jats:inline-formula>-dependence of the ridge or the away-side double-peak structure. The dropping ridge with
<jats:inline-formula>
<jats:tex-math><?CDATA $ \phi_{s} $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M10.jpg" xlink:type="simple" />
</jats:inline-formula> could be attributed to a
<jats:inline-formula>
<jats:tex-math><?CDATA $ \phi_{s} $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M11.jpg" xlink:type="simple" />
</jats:inline-formula>-dependent elliptic anisotropy; however, the physics mechanism of the ridge remains an open question. Even with a
<jats:inline-formula>
<jats:tex-math><?CDATA $ \phi_{s} $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M12.jpg" xlink:type="simple" />
</jats:inline-formula>-dependent elliptic flow, the away-side correlation structure is robust. These results, with extensive systematic studies of the dihadron correlations as a function of
<jats:inline-formula>
<jats:tex-math><?CDATA $ \phi_{s} $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M13.jpg" xlink:type="simple" />
</jats:inline-formula>, trigger and associated particle
<jats:inline-formula>
<jats:tex-math><?CDATA $ p_{T} $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M14.jpg" xlink:type="simple" />
</jats:inline-formula>, and the pseudo-rapidity range
<jats:inline-formula>
<jats:tex-math><?CDATA $ \Delta\eta $?></jats:tex-math>
<jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="cpc_45_4_044002_M15.jpg" xlink:type="simple" />
</jats:inline-formula>, should provide stringent inputs to help understand the underlying physics mechanisms of jet-medium interactions in high energy nuclear collisions.
</jats:p>
作者机构：
[Deng, Weibing; Li, Wei; Zhang, Wenjun] Cent China Normal Univ, Key Lab Quark & Lepton Phys MOE, Wuhan 430079, Peoples R China.;[Deng, Weibing; Li, Wei; Zhang, Wenjun] Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.;[Li, Wei] Max Planck Inst Math Sci, Inselstr 22-26, D-04103 Leipzig, Germany.;[Zhang, Wenjun] Anhui Univ Chinese Med, Sch Med Informat Engn, Hefei 230012, Peoples R China.
通讯机构：
[Weibing Deng] K;Key Laboratory of Quark and Lepton Physics (MOE) and Institute of Particle Physics, Central China Normal University, Wuhan 430079, China
关键词：
Cycle nodes ratio;Depth first search;Giant component;Network classification
摘要：
Cycles, which can be found in many different kinds of networks, make the problems more intractable, especially when dealing with dynamical processes on networks. On the contrary, tree networks in which no cycle exists, are simplifications and usually allow for analyticity. There lacks a quantity, however, to tell the ratio of cycles which determines the extent of network being close to tree networks. Therefore we introduce the term Cycle Nodes Ratio (CNR) to describe the ratio of number of nodes belonging to cycles to the number of total nodes, and provide an algorithm to calculate CNR. CNR is studied in both network models and real networks. The CNR remains unchanged in different sized Erdos--R & eacute;nyi (ER) networks with the same average degree, and increases with the average degree, which yields a critical turning point. The approximate analytical solutions of CNR in ER networks are given, which fits the simulations well. Furthermore, the difference between CNR and two-core ratio (TCR) is analyzed. The critical phenomenon is explored by analysing the giant component of networks. We compare the CNR in network models and real networks, and find the latter is generally smaller. Combining the coarse graining method can distinguish the CNR structure of networks with high average degree. The CNR is also applied to four different kinds of transportation networks and fungal networks, which give rise to different zones of effect. It is interesting to see that CNR is very useful in network recognition of machine learning. (c) 2021 Elsevier B.V. All rights reserved.
期刊：
Science Journal of Education,2021年9(3):104-109 ISSN：2329-0900
作者机构：
[Wei Li; Zhi Xin Huang; Ming Hai Wu; Qi Tian] College of Physical Science and Technology, Central China Normal University, Wuhan, China
关键词：
Classroom Response System;Deep Learning;Middle School Physics Review
摘要：
Deep learning is an important research topic in the field of educational technology, the Horizon Report released by the New Media Alliance in 2017 points out that deep learning is a key direction in promoting educational development and reform. This research is based on the CRS review discussion teaching mode, applying classroom response system in interactive teaching environment, in order to promot students' deep learning. Aiming at the problems of single teaching method, simple classroom communication, fuzzy monitoring of learning situation and low efficiency of teaching goal in junior high school physics review course, this paper takes junior high school physics review course as an example, discusses the teaching efficiency of CRS-based review course qualitatively, and uses comparative experiment to analyze the difference between teaching effect, students' deep learning achievement degree and learning type. The results show that, Based on the CRS junior high school physics review class in-depth learning model can improve students' academic performance, enhance the degree of achievement of students' deep learning. Therefore, we should actively explore the deep integration path of CRS and physics teaching, give full play to the advantages of CRS in focusing on the core issues in the classroom, evaluation of innovative learning process to improve the effectiveness of physics teaching.
作者机构：
[Shen, Jianmin; Li, Wei; Deng, Shengfeng; Zhang, Tao] Cent China Normal Univ, Key Lab Quark & Lepton Phys MOE, Wuhan 430079, Peoples R China.;[Shen, Jianmin; Li, Wei; Deng, Shengfeng; Zhang, Tao] Cent China Normal Univ, Inst Particle Phys, Wuhan 430079, Peoples R China.;[Li, Wei] Max Planck Inst Math Sci, D-04103 Leipzig, Germany.
摘要：
Machine learning (ML) has been well applied to studying equilibrium phase transition models by accurately predicating critical thresholds and some critical exponents. Difficulty will be raised, however, for integrating ML into nonequilibrium phase transitions. The extra dimension in a given nonequilibrium system, namely time, can greatly slow down the procedure toward the steady state. In this paper we find that by using some simple techniques of ML, non-steady-state configurations of directed percolation (DP) suffice to capture its essential critical behaviors in both ( 1+1 ) and ( 2+1 ) dimensions. With the supervised learning method, the framework of our binary classification neural networks can identify the phase transition threshold, as well as the spatial and temporal correlation exponents. The characteristic time tc , specifying the transition from active phases to absorbing ones, is also a major product of the learning. Moreover, we employ the convolutional autoencoder, an unsupervised learning technique, to extract dimensionality reduction representations and cluster configurations of ( 1+1 ) bond DP. It is quite appealing that such a method can yield a reasonable estimation of the critical point.
