By introducing the concepts and methods of machine learning, the traditional tensor network algorithm is upgraded to the machine learning tensor network algorithm in order to obtain more accurate results in complex systems. Under the context of the virtual time evolution tensor network algorithm established in the graded projection entangled pair state (gPEPS) representation of the two-dimensional strongly correlated electron infinite fermion system based on machine learning, this paper mainly investigates the quantum phase diagram, quantum phase transition and quantum entanglement of the two-dimensional t-J model related to high temperature superconductivity through the ground state energy. The research results showed that the t-J model has a phase separation line between the Heisenberg antiferromagnetic state with no vacancies and the state rich in vacancies, with the upper critical point at Jc=0.95t and the lower critical point at Jc=3.45t. It was found that at J=0.4t, in addition to the potential long-range ferromagnetic state, four different phases were composed in the different doping uniform regions: one wasis the superconducting phase with coexistence of charge and spin density waves and px(py) wave symmetry, one wasis the mixed superconducting phase with d+s wave superconducting pairing and spin triplet px(py) wave superconducting symmetry in the antiferromagnetic background, one wasis the superconducting phase with extended s-wave symmetry, and the other one iswas the superconducting phase with px(py) wave symmetry in the ferromagnetic background.
By introducing the concepts and methods of machine learning, the traditional tensor network algorithm is upgraded to the machine learning tensor network algorithm in order to obtain more accurate results in complex systems. Under the context of the virtual time evolution tensor network algorithm established in the graded projection entangled pair state (gPEPS) representation of the two-dimensional strongly correlated electron infinite fermion system based on machine learning, this paper mainly investigates the quantum phase diagram, quantum phase transition and quantum entanglement of the two-dimensional t-J model related to high temperature superconductivity through the ground state energy. The research results showed that the t-J model has a phase separation line between the Heisenberg antiferromagnetic state with no vacancies and the state rich in vacancies, with the upper critical point at Jc=0.95t and the lower critical point at Jc=3.45t. It was found that at J=0.4t, in addition to the potential long-range ferromagnetic state, four different phases were composed in the different doping uniform regions: one wasis the superconducting phase with coexistence of charge and spin density waves and px(py) wave symmetry, one wasis the mixed superconducting phase with d+s wave superconducting pairing and spin triplet px(py) wave superconducting symmetry in the antiferromagnetic background, one wasis the superconducting phase with extended s-wave symmetry, and the other one iswas the superconducting phase with px(py) wave symmetry in the ferromagnetic background.
The ground-state phase diagram of the two-dimensional t-J model is investigated in the context of the tensor network algorithm in terms of the graded Projected Entangled-Pair State representation of the ground-state wave functions. There is a line of phase separation between the Heisenberg anti-ferromagnetic state without hole and a hole-rich state. For both J=0.4t and J=0.8t, a systematic computation is performed to identify all the competing ground states for various dopings. It is found that, besides a possible Nagaoka's ferromagnetic state, the homogeneous regime consists of four different phases: one phase with charge and spin density wave order coexisting with a p_x (p_y)-wave superconducting state, one phase with the symmetry mixing of d+s-wave superconductivity in the spin-singlet channel and p_x (p_y)-wave superconductivity in the spin-triplet channel in the presence of an anti-ferromagnetic background, one superconducting phase with extended s-wave symmetry, and one superconducting phase with p_x (p_y)-wave symmetry in a ferromagnetic background.
The Tensor Network (TN) algorithm for strongly related electronic quantum lattice systems has become the uppermost algorithm for studying strongly related electronic quantum lattice systems in numerical computation. Focusing on the TN algorithm based on 2-D spin systems, this paper discusses the theoretical principles of the Projected Entangled Pair States (PEPS) algorithm and its implementation process. It also explores the infinite PEPS (iPEPS) algorithm for simulating the spin-1/2 infinite 2-D square lattice quantum XYX model. The critical phenomena of the two-dimensional XYX quantum system are analyzed from the perspectives of ground-state fidelity and order parameters, and the quantum critical point is identified. It is anticipated that with the further development of TN algorithms, the single-site fidelity theory of quantum phase transitions (QPTs) can be applied to various quantum many-body systems in condensed matter physics, thereby deepening our understanding of the extremely rich states of matter in the quantum world. Research indicates that quantum information theory provides a universal method for the study of QPTs, which can identify several ground-state fidelities and order parameters through the ground-state wave function of the system, several ground state fidelity and local order parameters can be identified to fully describe some quantum phase transition points, greatly enriching the traditional Landau theory of symmetry breaking QPTs.
Studying quantum phase transitions through order parameters is a traditional method, but studying phase transitions by machine learning is a brand new field. The ability of machine learning to classify, identify, or interpret massive data sets may provide physicists with similar analyses of the exponentially large data sets embodied in the Hilbert space of quantum many-body system. In this work, we propose a method of using unsupervised learning algorithm of the Gaussian mixture model to classify the state vectors of the J1-J2 antiferromagnetic Heisenberg spin chain system, then the supervised learning algorithm of the convolutional neural network is used to identify the classification point given by the unsupervised learning algorithm, and the cross-validation method is adopted to verify the learning effect. Using this method, we study the J1-J2 Heisenberg spin chain system with chain length N = 8, 10, 12, 16 and obtain the same conclusion. The first order phase transition point of J1-J2 antiferromagnetic Heisenberg spin chain system can be accurately found from the ground state vector, but the infinite order phase transition point cannot be found from the ground state vector. The first order and the infinite order phase transition point can be found from the first excited state vector, which indirectly shows that the first excited state may contain more information than the ground state of J1-J2 antiferromagnetic Heisenberg spin chain system. The visualization of the state vector shows the reliability of the machine learning algorithm, which can extract the feature information from the state vector. The result reveals that the machine learning techniques can directly find some possible phase transition points from a large set of state vectorwithout prior knowledge of the energy or locality conditions of the Hamiltonian, which may assists us in studying unknown systems. Supervised learning can verify the phase transition points given by unsupervised learning, thereby indicating that we can discover some useful information about unknown systems only through machine learning techniques. Machine learning techniques can be a basic research tool in strong quantum-correlated systems, and it can be adapted to more complex systems, which can help us dig up hidden information.
An algorithm to find a graded Projected Entangled-Pair State representation of the ground state wave functions is developed for translationally invariant strongly correlated electronic systems on infinite-size lattices in two spatial dimensions. It is tested for the two-dimensional t-J model at and away from half filling, with truncation dimensions up to 6. We are able to locate a line of phase separation, which qualitatively agrees with the results based on the high-temperature expansions. We find that the model exhibits an extended s-wave superconductivity for J=0.4t at quarter filling. However, we emphasize that the currently accessible truncation dimensions are not large enough, so it is necessary to incorporate the symmetry of the system into the algorithm, in order to achieve results with higher precision.
In the last few decades, quantum criticality in itinerant electron systems has become a central focus of condensed matter physics. On the one hand, it represents a candidate mechanism for high-temperature superconductivity. On the other hand, it can lead to a breakdown of Fermi liquid theory. The formation of a comprehensive understanding of metallic quantum criticality has, however, been significantly hampered by the fact that in many-fermion systems, fluctuations of a critical order parameter can couple to extensive gapless modes on a finite Fermi surface. This interplay, while giving rise to intriguing physical phenomena, leads to strong electronic correlations, which are notoriously difficult to handle by analytic methods. In this thesis, we investigate metallic quantum criticality by means of large-scale quantum Monte Carlo simulations and contribute unbiased, rigorous results to the discussion. Focusing on antiferromagnetic spin-density wave ordering, we present the numerically exact phase diagram of a spin-fermion model whose solution has so far been out of reach. In particular, we highlight the emergence of high-temperature d-wave superconductivity and rigorously establish the quantum critical properties of the antiferromagnetic quantum critical point. Combining the Monte Carlo method with a quantum loop topography approach, we demonstrate that important features of quantum critical metals can be autonomously identified by machine learning of current-current correlations. This allows us to analyze the electronic transport characteristics of two quantum critical metals, including the spin-fermion model, over a large parameter range and leads to the identification of extended non-Fermi liquid regimes in their respective phase diagrams.
Tensor network (TN) algorithms for strongly correlated electron lattice systems have become one of the most important numerical methods for studying such systems. Based on the tensor network algorithm for spin ladder systems, the study of quantum phase transitions (QPTs) in spin ladder systems through string order parameters has shown that quantum information theory provides a universal method for studying quantum phase transitions: without knowing the local or non-local order parameters of the system, or whether local or non-local order parameters exist, the ground-state wave function of the system can be obtained. If the ground-state lacks the traditional Landau long-range order and thus has no local order parameters, or if several non-local order parameters are found through other means, some topological quantum phases can be completely described, thereby greatly enriching the traditional Landau symmetry-breaking phase transition theory.