Deep neural networks(DNNs)and auto differentiation have been widely used in computational physics to solve variational problems.When a DNN is used to represent the wave function and solve quantum many-body problems us...Deep neural networks(DNNs)and auto differentiation have been widely used in computational physics to solve variational problems.When a DNN is used to represent the wave function and solve quantum many-body problems using variational optimization,various physical constraints have to be injected into the neural network by construction to increase the data and learning efficiency.We build the unitary constraint to the variational wave function using a monotonic neural network to represent the cumulative distribution function(CDF)F(x)=ʃ^(x)_(-∞)Ψ*Ψdx',.Using this constrained neural network to represent the variational wave function,we solve Schrodinger equations using auto-differentiation and stochastic gradient descent(SGD)by minimizing the violation of the trial wave function(x)to the Schrodinger equation.For several classical problems in quantum mechanics,we obtain their ground state wave function and energy with very low errors.The method developed in the present paper may pave a new way for solving nuclear many-body problems in the future.展开更多
In recent years, machine learning(ML) techniques have emerged as powerful tools for studying many-body complex systems, and encompassing phase transitions in various domains of physics. This mini review provides a con...In recent years, machine learning(ML) techniques have emerged as powerful tools for studying many-body complex systems, and encompassing phase transitions in various domains of physics. This mini review provides a concise yet comprehensive examination of the advancements achieved in applying ML to investigate phase transitions, with a primary focus on those involved in nuclear matter studies.展开更多
In relativistic heavy ion collisions,the fluctuations of initial entropy density convert to the correlations of final state hadrons in momentum space through the collective expansion of strongly interacting QCD matter...In relativistic heavy ion collisions,the fluctuations of initial entropy density convert to the correlations of final state hadrons in momentum space through the collective expansion of strongly interacting QCD matter.Using a(3+1)D viscous hydrodynamic program,CL Visc,we consider whether the nuclear structure,which provides initial state fluctuations as well as correlations,can affect the final state of heavy ion collisions,and whether one can find signals of α cluster structures in oxygen using final state observables in ^(16)O+ ^(16)O collisions at the CERN Large Hadron Collider.For the initial nucleon distributions in oxygen nuclei,we compare three different configurations,a tetrahedral structure with four-α clusters,the deformed Woods-Saxon distribution,and a spherical symmetric Woods-Saxon distribution.Our results show that the charged multiplicity as a function of centrality and the elliptic flow at the most central collisions using the four-α structure differs from those with the Woods-Saxon and deformed Woods-Saxon distributions,which may help to identify α clustering structures in oxygen nuclei.展开更多
基金Supported by the National Natural Science Foundation of China(12035006,12075098)the Natural Science Foundation of Hubei Province(2019CFB563)+1 种基金the Hubei Province Department of Education(D20201108)Hubei Province Department of Science and Technology(2021BLB171)。
文摘Deep neural networks(DNNs)and auto differentiation have been widely used in computational physics to solve variational problems.When a DNN is used to represent the wave function and solve quantum many-body problems using variational optimization,various physical constraints have to be injected into the neural network by construction to increase the data and learning efficiency.We build the unitary constraint to the variational wave function using a monotonic neural network to represent the cumulative distribution function(CDF)F(x)=ʃ^(x)_(-∞)Ψ*Ψdx',.Using this constrained neural network to represent the variational wave function,we solve Schrodinger equations using auto-differentiation and stochastic gradient descent(SGD)by minimizing the violation of the trial wave function(x)to the Schrodinger equation.For several classical problems in quantum mechanics,we obtain their ground state wave function and energy with very low errors.The method developed in the present paper may pave a new way for solving nuclear many-body problems in the future.
基金partially supported by the National Natural Science Foundation of China(Grant Nos. 11890710, 11890714, and 12147101)the BMBF funded KISS consortium (Grant No. 05D23RI1) in the ErUM-Data action plan。
文摘In recent years, machine learning(ML) techniques have emerged as powerful tools for studying many-body complex systems, and encompassing phase transitions in various domains of physics. This mini review provides a concise yet comprehensive examination of the advancements achieved in applying ML to investigate phase transitions, with a primary focus on those involved in nuclear matter studies.
基金Supported in part by the National Natural Science Foundation of China (12075098, 12147101, 11875066, 11861131009)Computations Were Performed at the Nuclear Science Computer Center at the CCNU (NSC3)。
文摘In relativistic heavy ion collisions,the fluctuations of initial entropy density convert to the correlations of final state hadrons in momentum space through the collective expansion of strongly interacting QCD matter.Using a(3+1)D viscous hydrodynamic program,CL Visc,we consider whether the nuclear structure,which provides initial state fluctuations as well as correlations,can affect the final state of heavy ion collisions,and whether one can find signals of α cluster structures in oxygen using final state observables in ^(16)O+ ^(16)O collisions at the CERN Large Hadron Collider.For the initial nucleon distributions in oxygen nuclei,we compare three different configurations,a tetrahedral structure with four-α clusters,the deformed Woods-Saxon distribution,and a spherical symmetric Woods-Saxon distribution.Our results show that the charged multiplicity as a function of centrality and the elliptic flow at the most central collisions using the four-α structure differs from those with the Woods-Saxon and deformed Woods-Saxon distributions,which may help to identify α clustering structures in oxygen nuclei.
