Drawing Phase Diagrams of Random Quantum Systems by Deep Learning the Wave Functions

Applications of neural networks to condensed matter physics are becoming popular and beginning to be well accepted. Obtaining and representing the ground and excited state wave functions are examples of such applications. Another application is analyzing the wave functions and determining their quantum phases. Here, we review the recent progress of using the multilayer convolutional neural network, so-called deep learning, to determine the quantum phases in random electron systems. After training the neural network by the supervised learning of wave functions in restricted parameter regions in known phases, the neural networks can determine the phases of the wave functions in wide parameter regions in unknown phases; hence, the phase diagrams are obtained. We demonstrate the validity and generality of this method by drawing the phase diagrams of two- and higher dimensional Anderson metal-insulator transitions and quantum percolations as well as disordered topological systems such as three-dimensional topological insulators and Weyl semimetals. Both real-space and Fourier space wave functions are analyzed. The advantages and disadvantages over conventional methods are discussed.

[1]  Zohar Ringel,et al.  Learning Curves for Deep Neural Networks: A Gaussian Field Theory Perspective , 2019, ArXiv.

[2]  F. Alet,et al.  Neural network setups for a precise detection of the many-body localization transition: Finite-size scaling and limitations , 2019, Physical Review B.

[3]  Naftali Tishby,et al.  Machine learning and the physical sciences , 2019, Reviews of Modern Physics.

[4]  Mats Granath,et al.  Unsupervised learning using topological data augmentation , 2019, Physical Review Research.

[5]  Binghai Yan,et al.  Active learning algorithm for computational physics , 2019, Physical Review Research.

[6]  Firas Hamze,et al.  Learning to find order in disorder , 2019, Journal of Statistical Mechanics: Theory and Experiment.

[7]  Weinan E,et al.  Deep neural network for the dielectric response of insulators , 2020 .

[8]  Hieu Chi Dam,et al.  A regression-based feature selection study of the Curie temperature of transition-metal rare-earth compounds: prediction and understanding , 2017 .

[9]  Ankit B. Patel,et al.  Deep learning-enhanced variational Monte Carlo method for quantum many-body physics , 2019, Physical Review Research.

[10]  Feng Liu,et al.  An Accurate and Transferable Machine-Learning Interatomic Potential for Silicon , 2019 .

[11]  Takahiro Sagawa,et al.  Learning the best thermoelectric nanoscale heat engines through evolving network topology , 2019, 1908.04866.

[12]  J. S. Bechtel,et al.  Machine learning the density functional theory potential energy surface for the inorganic halide perovskite CsPbBr3 , 2019, Physical Review B.

[13]  Hao Wu,et al.  Boltzmann generators: Sampling equilibrium states of many-body systems with deep learning , 2018, Science.

[14]  Hong Zhao,et al.  A New Strategy in Applying the Learning Machine to Study Phase Transitions. , 2019, 1901.00774.

[15]  M. Chung,et al.  Deep learning of topological phase transitions from entanglement aspects , 2019, Physical Review B.

[16]  Hiroshi Sawada,et al.  Machine-learning-assisted thin-film growth: Bayesian optimization in molecular beam epitaxy of SrRuO3 thin films , 2019, APL Materials.

[17]  J. Ignacio Cirac,et al.  From Probabilistic Graphical Models to Generalized Tensor Networks for Supervised Learning , 2018, IEEE Access.

[18]  Richard Kenway,et al.  Vulnerability of Deep Learning , 2018, ArXiv.

[19]  M. Süzen,et al.  Spectral Ergodicity in Deep Learning Architectures via Surrogate Random Matrices. , 2017 .

[20]  The bulletin of mathematical biophysics , 2005, Protoplasma.

[21]  Levent Sagun,et al.  Scaling description of generalization with number of parameters in deep learning , 2019, Journal of Statistical Mechanics: Theory and Experiment.

[22]  N. L. Holanda,et al.  Machine learning topological phases in real space , 2019, Physical Review B.

[23]  Anran Yuan,et al.  Machine Learning Magnetic Parameters from Spin Configurations , 2020, Advanced science.

[24]  Chiho Kim,et al.  Assessing and Improving Machine Learning Model Predictions of Polymer Glass Transition Temperatures , 2019 .

[25]  Conrad W. Rosenbrock,et al.  Machine-learned interatomic potentials for alloys and alloy phase diagrams , 2019, npj Computational Materials.

[26]  Toshiaki Koike-Akino,et al.  Deep Neural Network Inverse Design of Integrated Nanophotonic Devices. , 2018, 1809.03555.

[27]  Andrey Kardashin,et al.  Quantum Machine Learning Tensor Network States , 2018, Frontiers in Physics.

[28]  David J. Schwab,et al.  Supervised Learning with Tensor Networks , 2016, NIPS.

[29]  Kenta Hongo,et al.  Machine‐Learning Clustering Technique Applied to Powder X‐Ray Diffraction Patterns to Distinguish Compositions of ThMn 12 ‐Type Alloys , 2020, Advanced Theory and Simulations.

[30]  S. Redner,et al.  Introduction To Percolation Theory , 2018 .

[31]  Timo Hyart,et al.  Machine learning assisted measurement of local topological invariants , 2019, 1901.03346.

[32]  Liang Li,et al.  A Deep Learning Model for Atomic Structures Prediction Using X-ray Absorption Spectroscopic Data. , 2019 .

[33]  Nicolas Macris,et al.  The committee machine: computational to statistical gaps in learning a two-layers neural network , 2018, NeurIPS.

[34]  Qiang Du,et al.  Using a machine learning approach to determine the space group of a structure from the atomic pair distribution function. , 2019, Acta crystallographica. Section A, Foundations and advances.

[35]  Shun-ichi Amari,et al.  Universal statistics of Fisher information in deep neural networks: mean field approach , 2018, AISTATS.

[36]  Stefano Sanvito,et al.  Predicting the Curie temperature of ferromagnets using machine learning , 2019, Physical Review Materials.

[37]  Andrey Kardashin,et al.  Machine Learning Phase Transitions with a Quantum Processor , 2019, Physical Review A.

[38]  Arash Nikoubashman,et al.  BoltzmaNN: Deriving effective pair potentials and equations of state using neural networks , 2019, 1908.02448.

[39]  A. Trombettoni,et al.  Detecting composite orders in layered models via machine learning , 2020 .

[40]  R. Oliveira,et al.  A mean-field limit for certain deep neural networks , 2019, 1906.00193.

[41]  Grant M. Rotskoff,et al.  Neural Networks as Interacting Particle Systems: Asymptotic Convexity of the Loss Landscape and Universal Scaling of the Approximation Error , 2018, ArXiv.

[42]  Phan-Minh Nguyen,et al.  Mean Field Limit of the Learning Dynamics of Multilayer Neural Networks , 2019, ArXiv.

[43]  Felix Hill,et al.  Applying machine learning optimization methods to the production of a quantum gas , 2019, Mach. Learn. Sci. Technol..

[44]  David J. Schwab,et al.  An exact mapping between the Variational Renormalization Group and Deep Learning , 2014, ArXiv.

[45]  K. Nordlund,et al.  Machine-learning interatomic potential for radiation damage and defects in tungsten , 2019, Physical Review B.

[46]  Klaus-Robert Muller,et al.  Construction of Machine Learned Force Fields with Quantum Chemical Accuracy: Applications and Chemical Insights , 2020 .

[47]  Lu Wei,et al.  Entanglement area law for shallow and deep quantum neural network states , 2019, New Journal of Physics.

[48]  C. Chrysostomou,et al.  The critical temperature of the 2D-Ising model through deep learning autoencoders , 2020, The European Physical Journal B.

[49]  K. Sagiyama,et al.  Machine learning materials physics: Deep neural networks trained on elastic free energy data from martensitic microstructures predict homogenized stress fields with high accuracy , 2019, 1901.00524.

[50]  E. O. Kiktenko,et al.  Revealing quantum chaos with machine learning , 2019, Physical Review B.

[51]  Kipton Barros,et al.  Machine-learning-assisted insight into spin ice Dy2Ti2O7 , 2019, Nature Communications.

[52]  Takuya Matsumoto,et al.  Classifying Topological Charge in SU(3) Yang-Mills Theory with Machine Learning , 2019, ArXiv.

[53]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[54]  Hong-Ye Hu,et al.  Machine learning holographic mapping by neural network renormalization group , 2019, Physical Review Research.

[55]  Jonathan Schmidt,et al.  Machine Learning the Physical Nonlocal Exchange-Correlation Functional of Density-Functional Theory. , 2019, The journal of physical chemistry letters.

[56]  Yuki Nagai,et al.  Self-learning Monte Carlo method with Behler-Parrinello neural networks , 2018, Physical Review B.

[57]  Chris Wolverton,et al.  Developing an improved crystal graph convolutional neural network framework for accelerated materials discovery , 2019, 1906.05267.

[58]  Atsuto Seko,et al.  Descriptors for Machine Learning of Materials Data , 2017, 1709.01666.

[59]  Tota Nakamura,et al.  Machine learning as an improved estimator for magnetization curve and spin gap , 2019, Scientific Reports.

[60]  Zhaocheng Liu,et al.  Simulating the Ising Model with a Deep Convolutional Generative Adversarial Network , 2017, 1710.04987.

[61]  Gang Su,et al.  Quantum Compressed Sensing with Unsupervised Tensor Network Machine Learning , 2019, ArXiv.

[62]  F. Haake.  Quantum signatures of chaos , 1991 .

[63]  Yuma Iwasaki,et al.  Materials development by interpretable machine learning , 2019, 1903.02175.

[64]  Michele Ceriotti,et al.  Machine Learning of Atomic-Scale Properties Based on Physical Principles , 2019, Handbook of Materials Modeling.

[65]  Arun Mannodi-Kanakkithodi,et al.  Machine-learned impurity level prediction for semiconductors: the example of Cd-based chalcogenides , 2019, npj Computational Materials.

[66]  Emanuele Boattini,et al.  Unsupervised learning for local structure detection in colloidal systems. , 2019, The Journal of chemical physics.

[67]  Carlo Baldassi,et al.  Properties of the Geometry of Solutions and Capacity of Multilayer Neural Networks with Rectified Linear Unit Activations. , 2019, Physical review letters.

[68]  Kenny Choo,et al.  Two-dimensional frustrated J1−J2 model studied with neural network quantum states , 2019, Physical Review B.

[69]  C. Bény.  Deep learning and the renormalization group , 2013, 1301.3124.

[70]  David J. Schwab,et al.  Supervised Learning with Quantum-Inspired Tensor Networks , 2016, ArXiv.

[71]  Ryusuke Jinno,et al.  Machine learning for bounce calculation , 2018, 1805.12153.

[72]  Xin Qian,et al.  Thermal conductivity modeling using machine learning potentials: application to crystalline and amorphous silicon , 2019, Materials Today Physics.

[73]  Chahan M. Kropf,et al.  Deep neural network for x-ray photoelectron spectroscopy data analysis , 2019, Mach. Learn. Sci. Technol..

[74]  Simon Trebst,et al.  Quantum phase recognition via unsupervised machine learning , 2017, 1707.00663.

[75]  Grigory Bednik,et al.  Probing topological properties of a three-dimensional lattice dimer model with neural networks , 2019, Physical Review B.

[76]  Han-qing Shi,et al.  Neural-Network Quantum State of Transverse-Field Ising Model , 2019, Communications in Theoretical Physics.

[77]  Jiaxin Wu,et al.  Finding Quantum Many-Body Ground States with Artificial Neural Network , 2019, 1906.11216.

[78]  Sergei V. Kalinin,et al.  Learning from imperfections: constructing phase diagrams from atomic imaging of fluctuations , 2018, 1806.07475.