Showing 1–50 of 95 results for author: Ohzeki, M

Replica bound for Ising spin glass models in one dimension

Authors: Manaka Okuyama, Masayuki Ohzeki

Abstract: The interpolation method is a powerful tool for rigorous analysis of mean-field spin glass models, both with and without dilution. In this study, we show that the interpolation method can be applied to Ising spin glass models in one dimension, such as a one-dimensional chain and a two-leg ladder. In one dimension, the replica symmetric (RS) cavity method is naturally expected to be rigorous for Is… ▽ More The interpolation method is a powerful tool for rigorous analysis of mean-field spin glass models, both with and without dilution. In this study, we show that the interpolation method can be applied to Ising spin glass models in one dimension, such as a one-dimensional chain and a two-leg ladder. In one dimension, the replica symmetric (RS) cavity method is naturally expected to be rigorous for Ising spin glass models. Using the interpolation method, we rigorously prove that the RS cavity method provides lower bounds on the quenched free energies of Ising spin glass models in one dimension at any finite temperature in the thermodynamic limit. △ Less

Submitted 21 June, 2024; originally announced June 2024.

Comments: 16 pages, 0 figure

Free energy equivalence between mean-field models and nonsparsely diluted mean-field models

Authors: Manaka Okuyama, Masayuki Ohzeki

Abstract: We studied nonsparsely diluted mean-field models that differ from sparsely diluted mean-field models, such as the Viana--Bray model. We prove that the free energy of nonsparsely diluted mean-field models coincides exactly with that of the corresponding mean-field models with different parameters in ferromagnetic and spin-glass models composed of any discrete spin $S$ in the thermodynamic limit. Ou… ▽ More We studied nonsparsely diluted mean-field models that differ from sparsely diluted mean-field models, such as the Viana--Bray model. We prove that the free energy of nonsparsely diluted mean-field models coincides exactly with that of the corresponding mean-field models with different parameters in ferromagnetic and spin-glass models composed of any discrete spin $S$ in the thermodynamic limit. Our results are a broad generalization of the results of a previous study [Bovier and Gayrard, J. Stat. Phys. 72, 643 (1993)], where the densely diluted mean-field ferromagnetic Ising model (diluted Curie--Weiss model) was analyzed rigorously, and it was proven that its free energy was exactly equivalent to that of the corresponding mean-field model (Curie--Weiss model) with different parameters. △ Less

Submitted 24 June, 2024; v1 submitted 19 June, 2024; originally announced June 2024.

Comments: 9 pages, 0 figure

Toward mean-field bound for critical temperature on Nishimori line

Authors: Manaka Okuyama, Masayuki Ohzeki

Abstract: The critical inverse temperature of the mean-field approximation establishes a lower bound of the true critical inverse temperature in a broad class of ferromagnetic spin models. This is referred to as the mean-field bound for the critical temperature. In this study, we explored the possibility of a corresponding mean-field bound for the critical temperature in Ising spin glass models with Gaussia… ▽ More The critical inverse temperature of the mean-field approximation establishes a lower bound of the true critical inverse temperature in a broad class of ferromagnetic spin models. This is referred to as the mean-field bound for the critical temperature. In this study, we explored the possibility of a corresponding mean-field bound for the critical temperature in Ising spin glass models with Gaussian randomness on the Nishimori line. On this line, the critical inverse temperature of the mean-field approximation is given by $β_{MF}^{NL}=\sqrt{1/z}$, where $z$ is the coordination number. Using the Griffiths inequalities on the Nishimori line, we proved that there is zero spontaneous magnetization in the high-temperature region $β< β_{MF}^{NL}/2$. In other words, the true critical inverse temperature $β_c^{NL}$ on the Nishimori line is always bounded by $β_c^{NL} \ge β_{MF}^{NL}/2$. Unfortunately, we have not succeeded in obtaining the corresponding mean-field bound $β_c^{NL} \ge β_{MF}^{NL}$ on the Nishimori line. △ Less

Submitted 23 June, 2024; v1 submitted 18 June, 2024; originally announced June 2024.

Comments: 9pages, 0 figure

Reconsideration of optimization for reduction of traffic congestion

Authors: Masayuki Ohzeki

Abstract: One of the most impressive applications of a quantum annealer was optimizing a group of Volkswagen to reduce traffic congestion using a D-Wave system. A simple formulation of a quadratic term was proposed to reduce traffic congestion. This quadratic term was useful for determining the shortest routes among several candidates. The original formulation produced decreases in the total lengths of car… ▽ More One of the most impressive applications of a quantum annealer was optimizing a group of Volkswagen to reduce traffic congestion using a D-Wave system. A simple formulation of a quadratic term was proposed to reduce traffic congestion. This quadratic term was useful for determining the shortest routes among several candidates. The original formulation produced decreases in the total lengths of car tours and traffic congestion. In this study, we reformulated the cost function with the sole focus on reducing traffic congestion. We then found a unique cost function for expressing the quadratic function with a dead zone and an inequality constraint. △ Less

Submitted 8 June, 2024; originally announced June 2024.

Comments: 2 pages

Phase transition in binary compressed sensing based on $L_{1}$-norm minimization

Authors: Mikiya Doi, Masayuki Ohzeki

Abstract: Compressed sensing is a signal processing scheme that reconstructs high-dimensional sparse signals from a limited number of observations. In recent years, various problems involving signals with a finite number of discrete values have been attracting attention in the field of compressed sensing. In particular, binary compressed sensing, which restricts signal elements to binary values $\{0, 1\}$,… ▽ More Compressed sensing is a signal processing scheme that reconstructs high-dimensional sparse signals from a limited number of observations. In recent years, various problems involving signals with a finite number of discrete values have been attracting attention in the field of compressed sensing. In particular, binary compressed sensing, which restricts signal elements to binary values $\{0, 1\}$, is the most fundamental and straightforward analysis subject in such problem settings. We evaluate the typical performance of noiseless binary compressed sensing based on $L_{1}$-norm minimization using the replica method, a statistical mechanical approach. We analyze a general setting where the elements of the observation matrix follow a Gaussian distribution, including a non-zero mean. We demonstrate that the biased observation matrix indicates more reconstruction success conditions in binary compressed sensing. Our results are consistent with the outcomes of several prior studies. △ Less

Submitted 27 May, 2024; originally announced May 2024.

Comments: 14 pages, 2 Figures

Application of time-series quantum generative model to financial data

Authors: Shun Okumura, Masayuki Ohzeki, Masaya Abe

Abstract: Despite proposing a quantum generative model for time series that successfully learns correlated series with multiple Brownian motions, the model has not been adapted and evaluated for financial problems. In this study, a time-series generative model was applied as a quantum generative model to actual financial data. Future data for two correlated time series were generated and compared with class… ▽ More Despite proposing a quantum generative model for time series that successfully learns correlated series with multiple Brownian motions, the model has not been adapted and evaluated for financial problems. In this study, a time-series generative model was applied as a quantum generative model to actual financial data. Future data for two correlated time series were generated and compared with classical methods such as long short-term memory and vector autoregression. Furthermore, numerical experiments were performed to complete missing values. Based on the results, we evaluated the practical applications of the time-series quantum generation model. It was observed that fewer parameter values were required compared with the classical method. In addition, the quantum time-series generation model was feasible for both stationary and nonstationary data. These results suggest that several parameters can be applied to various types of time-series data. △ Less

Submitted 20 May, 2024; originally announced May 2024.

Comments: 12 pages

Statistical Mechanics Calculations Using Variational Autoregressive Networks and Quantum Annealing

Authors: Yuta Tamura, Masayuki Ohzeki

Abstract: In statistical mechanics, computing the partition function is generally difficult. An approximation method using a variational autoregressive network (VAN) has been proposed recently. This approach offers the advantage of directly calculating the generation probabilities while obtaining a significantly large number of samples. The present study introduces a novel approximation method that employs… ▽ More In statistical mechanics, computing the partition function is generally difficult. An approximation method using a variational autoregressive network (VAN) has been proposed recently. This approach offers the advantage of directly calculating the generation probabilities while obtaining a significantly large number of samples. The present study introduces a novel approximation method that employs samples derived from quantum annealing machines in conjunction with VAN, which are empirically assumed to adhere to the Gibbs-Boltzmann distribution. When applied to the finite-size Sherrington-Kirkpatrick model, the proposed method demonstrates enhanced accuracy compared to the traditional VAN approach and other approximate methods, such as the widely utilized naive mean field. △ Less

Submitted 20 May, 2024; v1 submitted 30 April, 2024; originally announced April 2024.

Comments: 5pages

Storage Capacity Evaluation of the Quantum Perceptron using the Replica Method

Authors: Mitsuru Urushibata, Masayuki Ohzeki

Abstract: We investigate a quantum perceptron implemented on a quantum circuit using a repeat until method. We evaluate this from the perspective of capacity, one of the performance evaluation measures for perceptions. We assess a Gardner volume, defined as a volume of coefficients of the perceptron that can correctly classify given training examples using the replica method. The model is defined on the qua… ▽ More We investigate a quantum perceptron implemented on a quantum circuit using a repeat until method. We evaluate this from the perspective of capacity, one of the performance evaluation measures for perceptions. We assess a Gardner volume, defined as a volume of coefficients of the perceptron that can correctly classify given training examples using the replica method. The model is defined on the quantum circuit. Nevertheless, it is straightforward to assess the capacity using the replica method, which is a standard method in classical statistical mechanics. The reason why we can solve our model by the replica method is the repeat until method, in which we focus on the output of the measurements of the quantum circuit. We find that the capacity of a quantum perceptron is larger than that of a classical perceptron since the quantum one is simple but effectively falls into a highly nonlinear form of the activation function. △ Less

Submitted 30 April, 2024; v1 submitted 23 April, 2024; originally announced April 2024.

Comments: 12 pages, 2 figures

Solution space and storage capacity of fully connected two-layer neural networks with generic activation functions

Authors: Sota Nishiyama, Masayuki Ohzeki

Abstract: The storage capacity of a binary classification model is the maximum number of random input-output pairs per parameter that the model can learn. It is one of the indicators of the expressive power of machine learning models and is important for comparing the performance of various models. In this study, we analyze the structure of the solution space and the storage capacity of fully connected two-… ▽ More The storage capacity of a binary classification model is the maximum number of random input-output pairs per parameter that the model can learn. It is one of the indicators of the expressive power of machine learning models and is important for comparing the performance of various models. In this study, we analyze the structure of the solution space and the storage capacity of fully connected two-layer neural networks with general activation functions using the replica method from statistical physics. Our results demonstrate that the storage capacity per parameter remains finite even with infinite width and that the weights of the network exhibit negative correlations, leading to a 'division of labor'. In addition, we find that increasing the dataset size triggers a phase transition at a certain transition point where the permutation symmetry of weights is broken, resulting in the solution space splitting into disjoint regions. We identify the dependence of this transition point and the storage capacity on the choice of activation function. These findings contribute to understanding the influence of activation functions and the number of parameters on the structure of the solution space, potentially offering insights for selecting appropriate architectures based on specific objectives. △ Less

Submitted 20 April, 2024; originally announced April 2024.

Comments: 10+11 pages, 5 figures, 1 table

Duality analysis in symmetric group and its application to random tensor network model

Authors: Masayuki Ohzeki

Abstract: The Ising model is the simplest to describe many-body effects in classical statistical mechanics. Duality analysis leads to a critical point under several assumptions. The Ising model itself has $Z(2)$ symmetry. The basis of the duality analysis is a nontrivial relationship between low and high-temperature expansions. However, the discrete Fourier transformation finds the hidden relationship autom… ▽ More The Ising model is the simplest to describe many-body effects in classical statistical mechanics. Duality analysis leads to a critical point under several assumptions. The Ising model itself has $Z(2)$ symmetry. The basis of the duality analysis is a nontrivial relationship between low and high-temperature expansions. However, the discrete Fourier transformation finds the hidden relationship automatically. The duality analysis can be naturally generalized into the case with the degrees of freedom with $Z(q)$ symmetry and random spin systems. We further obtain the duality in a series of permutation models in the present study by considering the symmetric group $S_q$ and its Fourier transformation. The permutation model in the symmetric group is closely related to the random quantum circuits and random tensor network model, often discussed in the context of quantum computing and the holographic principle, a property of string theories and quantum gravity. We provide a systematic way by our duality analysis to analyze the phase transition in these models. △ Less

Submitted 25 June, 2024; v1 submitted 21 October, 2023; originally announced October 2023.

Comments: 9 pages

Individual subject evaluated difficulty of adjustable mazes generated using quantum annealing

Authors: Yuto Ishikawa, Takuma Yoshihara, Keita Okamura, Masayuki Ohzeki

Abstract: In this paper, the maze generation using quantum annealing is proposed. We reformulate a standard algorithm to generate a maze into a specific form of a quadratic unconstrained binary optimization problem suitable for the input of the quantum annealer. To generate more difficult mazes, we introduce an additional cost function $Q_{update}$ to increase the difficulty. The difficulty of the mazes was… ▽ More In this paper, the maze generation using quantum annealing is proposed. We reformulate a standard algorithm to generate a maze into a specific form of a quadratic unconstrained binary optimization problem suitable for the input of the quantum annealer. To generate more difficult mazes, we introduce an additional cost function $Q_{update}$ to increase the difficulty. The difficulty of the mazes was evaluated by the time to solve the maze of 12 human subjects. To check the efficiency of our scheme to create the maze, we investigated the time-to-solution of a quantum processing unit, classical computer, and hybrid solver. △ Less

Submitted 10 November, 2023; v1 submitted 9 September, 2023; originally announced September 2023.

Comments: 20pages, 15figures

Random Postprocessing for Combinatorial Bayesian Optimization

Authors: Keisuke Morita, Yoshihiko Nishikawa, Masayuki Ohzeki

Abstract: Model-based sequential approaches to discrete "black-box" optimization, including Bayesian optimization techniques, often access the same points multiple times for a given objective function in interest, resulting in many steps to find the global optimum. Here, we numerically study the effect of a postprocessing method on Bayesian optimization that strictly prohibits duplicated samples in the data… ▽ More Model-based sequential approaches to discrete "black-box" optimization, including Bayesian optimization techniques, often access the same points multiple times for a given objective function in interest, resulting in many steps to find the global optimum. Here, we numerically study the effect of a postprocessing method on Bayesian optimization that strictly prohibits duplicated samples in the dataset. We find the postprocessing method significantly reduces the number of sequential steps to find the global optimum, especially when the acquisition function is of maximum a posterior estimation. Our results provide a simple but general strategy to solve the slow convergence of Bayesian optimization for high-dimensional problems. △ Less

Submitted 27 December, 2023; v1 submitted 6 September, 2023; originally announced September 2023.

Comments: 10 pages, 4 figures

Journal ref: J. Phys. Soc. Jpn. 92, 123801 (2023)

Online calibration scheme for training restricted Boltzmann machines with quantum annealing

Authors: Takeru Goto, Masayuki Ohzeki

Abstract: We propose a scheme for calibrating the D-Wave quantum annealer's internal parameters to obtain well-approximated samples to train a restricted Boltzmann machine (RBM). Empirically, samples from the quantum annealer obey the Boltzmann distribution, making them suitable for RBM training. However, it is hard to obtain appropriate samples without compensation. Existing research often estimates intern… ▽ More We propose a scheme for calibrating the D-Wave quantum annealer's internal parameters to obtain well-approximated samples to train a restricted Boltzmann machine (RBM). Empirically, samples from the quantum annealer obey the Boltzmann distribution, making them suitable for RBM training. However, it is hard to obtain appropriate samples without compensation. Existing research often estimates internal parameters, such as the inverse temperature, for compensation. Our scheme utilizes samples for RBM training to estimate the internal parameters, enabling it to train a model simultaneously. Furthermore, we consider additional parameters beyond inverse temperature and demonstrate that they contribute to improving sample quality. We evaluate the performance of our scheme by comparing the Kullback-Leibler divergence of the obtained samples with classical Gibbs sampling. Our results indicate that our proposed scheme demonstrates performance on par with Gibbs sampling. In addition, the training results with our estimation scheme are better than those of the Contrastive Divergence algorithm, known as a standard training algorithm for RBM. △ Less

Submitted 19 July, 2023; originally announced July 2023.

Comments: 10 pages, 5 figures, 2 tables

Efficient Algorithm for Binary Quadratic Problem by Column Generation and Quantum Annealing

Authors: Sota Hirama, Masayuki Ohzeki

Abstract: We propose an efficient algorithm that combines column generation and quantum annealing to solve binary quadratic problems. Binary quadratic problems are difficult to solve because they are NP-hard. An attempt to solve binary quadratic problems efficiently by column generation has been studied, but it demands successively solving quadratic unconstrained binary optimization problems. We solve the b… ▽ More We propose an efficient algorithm that combines column generation and quantum annealing to solve binary quadratic problems. Binary quadratic problems are difficult to solve because they are NP-hard. An attempt to solve binary quadratic problems efficiently by column generation has been studied, but it demands successively solving quadratic unconstrained binary optimization problems. We solve the bottleneck by using quantum annealing or simulated annealing. Our results demonstrate a good approximate solution obtained in 2.7 to 1000 times shorter computational time to use column generation and quantum annealing to solve binary quadratic problems than the existing fast solver. △ Less

Submitted 12 July, 2023; originally announced July 2023.

Comments: 5 pages, 3 figures, 1 Table

Report number: 113002

Journal ref: J. Phys. Soc. Jpn. 92, No.11 (2023)

Kernel Learning by quantum annealer

Authors: Yasushi Hasegawa, Hiroki Oshiyama, Masayuki Ohzeki

Abstract: The Boltzmann machine is one of the various applications using quantum annealer. We propose an application of the Boltzmann machine to the kernel matrix used in various machine-learning techniques. We focus on the fact that shift-invariant kernel functions can be expressed in terms of the expected value of a spectral distribution by the Fourier transformation. Using this transformation, random Fou… ▽ More The Boltzmann machine is one of the various applications using quantum annealer. We propose an application of the Boltzmann machine to the kernel matrix used in various machine-learning techniques. We focus on the fact that shift-invariant kernel functions can be expressed in terms of the expected value of a spectral distribution by the Fourier transformation. Using this transformation, random Fourier feature (RFF) samples the frequencies and approximates the kernel function. In this paper, furthermore, we propose a method to obtain a spectral distribution suitable for the data using a Boltzmann machine. As a result, we show that the prediction accuracy is comparable to that of the method using the Gaussian distribution. We also show that it is possible to create a spectral distribution that could not be feasible with the Gaussian distribution. △ Less

Submitted 20 April, 2023; originally announced April 2023.

Comments: 10 pages, 5 figures, 1 table

Magnetic order in nanoscale gyroid networks

Authors: Ami S. Koshikawa, Justin Llandro, Masayuki Ohzeki, Shunsuke Fukami, Hideo Ohno, Naëmi Leo

Abstract: Three-dimensional magnetic metamaterials feature interesting phenomena that arise from a delicate interplay of material properties, local anisotropy, curvature, and connectivity. A particularly interesting magnetic lattice that combines these aspects is that of nanoscale gyroids, with a highly-interconnected chiral network with local three-connectivity reminiscent of three-dimensional artificial s… ▽ More Three-dimensional magnetic metamaterials feature interesting phenomena that arise from a delicate interplay of material properties, local anisotropy, curvature, and connectivity. A particularly interesting magnetic lattice that combines these aspects is that of nanoscale gyroids, with a highly-interconnected chiral network with local three-connectivity reminiscent of three-dimensional artificial spin ices. Here, we use finite-element micromagnetic simulations to elucidate the anisotropic behaviour of nanoscale nickel gyroid networks at applied fields and at remanence. We simplify the description of the micromagnetic spin states with a macrospin model to explain the anistropic global response, to quantify the extent of ice-like correlations, and to discuss qualitative features of the anisotropic magnetoresistance in the three-dimensional network. Our results demonstrate the large variability of the magnetic order in extended gyroid networks, which might enable future spintronic functionalities, including neuromorphic computing and non-reciprocal transport. △ Less

Submitted 23 May, 2024; v1 submitted 10 March, 2023; originally announced March 2023.

Comments: 10 pages, 6 figures

Journal ref: Physical Review B 108 024414 (2023)

Mean-field theory is exact for Ising spin glass models with Kac potential in non-additive limit on Nishimori line

Authors: Manaka Okuyama, Masayuki Ohzeki

Abstract: Recently, Mori [Phys. Rev. E 84, 031128 (2011)] has conjectured that the free energy of Ising spin glass models with the Kac potential in the non-additive limit, such as the power-law potential in the non-additive regime, is exactly equal to that of the Sherrington-Kirkpatrick model in the thermodynamic limit. In this study, we prove that his conjecture is true on the Nishimori line at any tempera… ▽ More Recently, Mori [Phys. Rev. E 84, 031128 (2011)] has conjectured that the free energy of Ising spin glass models with the Kac potential in the non-additive limit, such as the power-law potential in the non-additive regime, is exactly equal to that of the Sherrington-Kirkpatrick model in the thermodynamic limit. In this study, we prove that his conjecture is true on the Nishimori line at any temperature in any dimension. One of the key ingredients of the proof is the use of the Gibbs-Bogoliubov inequality on the Nishimori line. We also consider the case in which the probability distribution of the interaction is symmetric, where his conjecture is true at any temperature in one dimension but is an open problem in the low-temperature regime in two or more dimensions. △ Less

Submitted 30 January, 2023; originally announced January 2023.

Comments: 12pages, 0 figure

Journal ref: J. Phys. A: Math. Theor. 56, 325003 (2023)

Gibbs-Bogoliubov inequality on Nishimori line

Authors: Manaka Okuyama, Masayuki Ohzeki

Abstract: The Gibbs-Bogoliubov inequality states that the free energy of a system is always lower than that calculated by a trial function. In this study, we show that a counterpart of the Gibbs-Bogoliubov inequality holds on the Nishimori line for Ising spin-glass models with Gaussian randomness. Our inequality states that the quenched free energy of a system is always lower than that calculated using a qu… ▽ More The Gibbs-Bogoliubov inequality states that the free energy of a system is always lower than that calculated by a trial function. In this study, we show that a counterpart of the Gibbs-Bogoliubov inequality holds on the Nishimori line for Ising spin-glass models with Gaussian randomness. Our inequality states that the quenched free energy of a system is always lower than that calculated using a quenched trial function. The key component of the proof is the convexity of the pressure function $\mathbb{E}\left[\log Z_{} \right]$ with respect to the parameters along the Nishimori line, which differs from the conventional convexity with respect to the inverse temperature. When our inequality was applied to mean-field models, such as the Sherrington-Kirkpatrick model and $p$-spin model, the bound coincided with the replica-symmetric solution indicating that the equality holds. △ Less

Submitted 6 June, 2023; v1 submitted 25 August, 2022; originally announced August 2022.

Comments: 9 pages, 0 figure

Journal ref: J. Phys. Soc. Jpn. 92, 084002 (2023)

Comparing the effects of Boltzmann machines as associative memory in Generative Adversarial Networks between classical and quantum sampling

Authors: Mitsuru Urushibata, Masayuki Ohzeki, Kazuyuki Tanaka

Abstract: We investigate the quantum effect on machine learning (ML) models exemplified by the Generative Adversarial Network (GAN), which is a promising deep learning framework. In the general GAN framework the generator maps uniform noise to a fake image. In this study, we utilize the Associative Adversarial Network (AAN), which consists of a standard GAN and an associative memory. Further, we set a Boltz… ▽ More We investigate the quantum effect on machine learning (ML) models exemplified by the Generative Adversarial Network (GAN), which is a promising deep learning framework. In the general GAN framework the generator maps uniform noise to a fake image. In this study, we utilize the Associative Adversarial Network (AAN), which consists of a standard GAN and an associative memory. Further, we set a Boltzmann Machine (BM), which is an undirected graphical model that learns low-dimensional features extracted from a discriminator, as the memory. Owing to difficulty calculating the BM's log-likelihood gradient, it is necessary to approximate it by using the sample mean obtained from the BM, which has tentative parameters. To calculate the sample mean, a Markov Chain Monte Carlo (MCMC) is often used. In a previous study, this was performed using a quantum annealer device, and the performance of the "Quantum" AAN was compared to that of the standard GAN. However, its better performance than the standard GAN is not well understood. In this study, we introduce two methods to draw samples: classical sampling via MCMC and quantum sampling via quantum Monte Carlo (QMC) simulation, which is quantum simulation on the classical computer. Then, we compare these methods to investigate whether quantum sampling is advantageous. Specifically, calculating the discriminator loss, the generator loss, inception score and Fréchet inception distance, we discuss the possibility of AAN. We show that the AANs trained by both MCMC and QMC are more stable during training and produce more varied images than the standard GANs. However, the results indicate no difference in sampling by QMC simulation compared to that by MCMC. △ Less

Submitted 29 March, 2022; originally announced March 2022.

Comments: 15 pages, 12 figures

Graph minor embedding of degenerate systems in quantum annealing

Authors: Naoki Maruyama, Masayuki Ohzeki, Kazuyuki Tanaka

Abstract: Quantum annealing, as currently implemented in hardware, cannot fairly sample all ground states. Graph minor embedding in a quantum annealer leads to biased sampling results. We demonstrate the influence of the embedding process on sampling results in a degenerate problem and analyze the details using perturbation theory. Our result also shows the relationship between the probabilities of ground s… ▽ More Quantum annealing, as currently implemented in hardware, cannot fairly sample all ground states. Graph minor embedding in a quantum annealer leads to biased sampling results. We demonstrate the influence of the embedding process on sampling results in a degenerate problem and analyze the details using perturbation theory. Our result also shows the relationship between the probabilities of ground states and the energy landscape between them. △ Less

Submitted 21 October, 2021; originally announced October 2021.

Comments: 2 pages, 3 figures

Combinatorial Black-box Optimization for Vehicle Design Problem

Authors: Ami S. Koshikawa, Masayuki Ohzeki, Masamichi J. Miyama, Kazuyuki Tanaka, Yusaku Yamashita, Johannes Stadler, Oliver Wick

Abstract: Black-box optimization minimizes an objective function without derivatives or explicit forms. Such an optimization method with continuous variables has been successful in the fields of machine learning and material science. For discrete variables, the Bayesian optimization of combinatorial structure (BOCS) is a powerful tool for solving black-box optimization problems. A surrogate model used in BO… ▽ More Black-box optimization minimizes an objective function without derivatives or explicit forms. Such an optimization method with continuous variables has been successful in the fields of machine learning and material science. For discrete variables, the Bayesian optimization of combinatorial structure (BOCS) is a powerful tool for solving black-box optimization problems. A surrogate model used in BOCS is the quadratic unconstrained binary optimization (QUBO) form. Because of the approximation of the objective function to the QUBO form in BOCS, BOCS can expand the possibilities of using D-Wave quantum annealers, which can generate near-optimal solutions of QUBO problems by utilizing quantum fluctuation. We demonstrate the use of BOCS and its variant for a vehicle design problem, which cannot be described in the QUBO form. As a result, BOCS and its variant slightly outperform the random search, which randomly calculates the objective function. △ Less

Submitted 1 October, 2021; originally announced October 2021.

Comments: 18pages, 23 figures

Stochastic dynamics without detailed balance condition connecting simple gradient method and Hamiltonian Monte Carlo

Authors: Akihisa Ichiki, Masayuki Ohzeki

Abstract: Sampling occupies an important position in theories of various scientific fields, and Markov chain Monte Carlo (MCMC) provides the most common technique of sampling. In the progress of MCMC, a huge number of studies have aimed the acceleration of convergence to the target distribution. Hamiltonian Monte Carlo (HMC) is such a variant of MCMC. In the recent development of MCMC, another approach base… ▽ More Sampling occupies an important position in theories of various scientific fields, and Markov chain Monte Carlo (MCMC) provides the most common technique of sampling. In the progress of MCMC, a huge number of studies have aimed the acceleration of convergence to the target distribution. Hamiltonian Monte Carlo (HMC) is such a variant of MCMC. In the recent development of MCMC, another approach based on the violation of the detailed balance condition has attracted much attention. Historically, these two approaches have been proposed independently, and their relationship has not been clearly understood. In this paper, the two approaches are seamlessly understood in the framework of generalized Monte Carlo method that violates the detailed balance condition. Furthermore we propose an efficient Monte Carlo method based on our framework. △ Less

Submitted 19 July, 2021; originally announced July 2021.

Comments: 2 figures

Mirror-symmetry-protected dynamical quantum phase transitions in topological crystalline insulators

Authors: Ryo Okugawa, Hiroki Oshiyama, Masayuki Ohzeki

Abstract: Dynamical quantum phase transitions (DQPTs) are topologically characterized in quantum quench dynamics in topological systems. In this paper, we study Loschmidt amplitudes and DQPTs in quantum quenches in mirror-symmetric topological phases. Based on the topological classification of mirror-symmetric insulators, we show that mirror symmetry creates symmetry-protected DQPTs. If mirror symmetry is p… ▽ More Dynamical quantum phase transitions (DQPTs) are topologically characterized in quantum quench dynamics in topological systems. In this paper, we study Loschmidt amplitudes and DQPTs in quantum quenches in mirror-symmetric topological phases. Based on the topological classification of mirror-symmetric insulators, we show that mirror symmetry creates symmetry-protected DQPTs. If mirror symmetry is present, topologically robust DQPTs can occur in quantum quenches, even in high-dimensional time-reversal invariant systems. Then, we also show that symmetry-protected DQPTs occur in quenches in two-dimensional chiral-symmetric systems with mirror symmetry. Mirror-symmetry-protected DQPTs can be easily captured by a reduced rate function. Moreover, we introduce dynamical topological order parameters for mirror-symmetry-protected DQPTs. Finally, we demonstrate DQPTs using lattice models for a time-reversal invariant topological crystalline insulator and a higher-order topological insulator. △ Less

Submitted 23 October, 2021; v1 submitted 26 May, 2021; originally announced May 2021.

Comments: 10 pages, 3 figures; Discussions and references added, and typos corrected

Journal ref: Phys. Rev. Research 3, 043064 (2021)

Benchmark test of Black-box optimization using D-Wave quantum annealer

Authors: Ami S. Koshikawa, Masayuki Ohzeki, Tadashi Kadowaki, Kazuyuki Tanaka

Abstract: In solving optimization problems, objective functions generally need to be minimized or maximized. However, objective functions cannot always be formulated explicitly in a mathematical form for complicated problem settings. Although several regression techniques infer the approximate forms of objective functions, they are at times expensive to evaluate. Optimal points of "black-box" objective func… ▽ More In solving optimization problems, objective functions generally need to be minimized or maximized. However, objective functions cannot always be formulated explicitly in a mathematical form for complicated problem settings. Although several regression techniques infer the approximate forms of objective functions, they are at times expensive to evaluate. Optimal points of "black-box" objective functions are computed in such scenarios, while effectively using a small number of clues. Recently, an efficient method by use of inference by sparse prior for a black-box objective function with binary variables has been proposed. In this method, a surrogate model was proposed in the form of a quadratic unconstrained binary optimization (QUBO) problem, and was iteratively solved to obtain the optimal solution of the black-box objective function. In the present study, we employ the D-Wave 2000Q quantum annealer, which can solve QUBO by driving the binary variables by quantum fluctuations. The D-Wave 2000Q quantum annealer does not necessarily output the ground state at the end of the protocol due to freezing effect during the process. We investigate effects from the output of the D-Wave quantum annealer in performing black-box optimization. We demonstrate a benchmark test by employing the sparse Sherrington-Kirkpatrick (SK) model as the black-box objective function, by introducing a parameter controlling the sparseness of the interaction coefficients. Comparing the results of the D-Wave quantum annealer to those of the simulated annealing (SA) and semidefinite programming (SDP), our results by the D-Wave quantum annealer and SA exhibit superiority in black-box optimization with SDP. On the other hand, we did not find any advantage of the D-Wave quantum annealer over the simulated annealing. As far as in our case, any effects by quantum fluctuation are not found. △ Less

Submitted 23 March, 2021; originally announced March 2021.

Comments: 22 pages, 4 figures

Journal ref: J. Phys. Soc. Jpn. 90, 064001 (2021)

Assessment of image generation by quantum annealer

Authors: Takehito Sato, Masayuki Ohzeki, Kazuyuki Tanaka

Abstract: Quantum annealing was originally proposed as an approach for solving combinatorial optimisation problems using quantum effects. D-Wave Systems has released a production model of quantum annealing hardware. However, the inherent noise and various environmental factors in the hardware hamper the determination of optimal solutions. In addition, the freezing effect in regions with weak quantum fluctua… ▽ More Quantum annealing was originally proposed as an approach for solving combinatorial optimisation problems using quantum effects. D-Wave Systems has released a production model of quantum annealing hardware. However, the inherent noise and various environmental factors in the hardware hamper the determination of optimal solutions. In addition, the freezing effect in regions with weak quantum fluctuations generates outputs approximately following a Gibbs--Boltzmann distribution at an extremely low temperature. Thus, a quantum annealer may also serve as a fast sampler for the Ising spin-glass problem, and several studies have investigated Boltzmann machine learning using a quantum annealer. Previous developments have focused on comparing the performance in the standard distance of the resulting distributions between conventional methods in classical computers and sampling by a quantum annealer. In this study, we focused on the performance of a quantum annealer as a generative model. To evaluate its performance, we prepared a discriminator given by a neural network trained on an a priori dataset. The evaluation results show a higher performance of quantum annealing compared with the classical approach for Boltzmann machine learning. △ Less

Submitted 15 March, 2021; originally announced March 2021.

Comments: 10 pages, 8 figures

Teacher-student learning for a binary perceptron with quantum fluctuations

Authors: Shunta Arai, Masayuki Ohzeki, Kazuyuki Tanaka

Abstract: We analysed the generalisation performance of a binary perceptron with quantum fluctuations using the replica method. An exponential number of local minima dominate the energy landscape of the binary perceptron. Local search algorithms often fail to identify the ground state of a binary perceptron. In this study, we considered the teacher-student learning method and computed the generalisation err… ▽ More We analysed the generalisation performance of a binary perceptron with quantum fluctuations using the replica method. An exponential number of local minima dominate the energy landscape of the binary perceptron. Local search algorithms often fail to identify the ground state of a binary perceptron. In this study, we considered the teacher-student learning method and computed the generalisation error of a binary perceptron with quantum fluctuations. Due to the quantum fluctuations, we can efficiently find robust solutions that have better generalisation performance than the classical model. We validated our theoretical results through quantum Monte Carlo simulations. We adopted the replica symmetry (RS) ansatz assumption and static approximation. The RS solutions are consistent with our simulation results, except for the relatively low strength of the transverse field and high pattern ratio. These deviations are caused by the violation of ergodicity and static approximation. After accounting for the deviation between the RS solutions and numerical results, the enhancement of generalisation performance with quantum fluctuations holds. △ Less

Submitted 22 April, 2021; v1 submitted 17 February, 2021; originally announced February 2021.

Comments: 12pages,5figures

Solving Inequality-Constrained Binary Optimization Problems on Quantum Annealer

Authors: Kouki Yonaga, Masamichi J. Miyama, Masayuki Ohzeki

Abstract: We propose a new method for solving binary optimization problems under inequality constraints using a quantum annealer. To deal with inequality constraints, we often use slack variables, as in previous approaches. When we use slack variables, we usually conduct a binary expansion, which requires numerous physical qubits. Therefore, the problem of the current quantum annealer is limited to a small… ▽ More We propose a new method for solving binary optimization problems under inequality constraints using a quantum annealer. To deal with inequality constraints, we often use slack variables, as in previous approaches. When we use slack variables, we usually conduct a binary expansion, which requires numerous physical qubits. Therefore, the problem of the current quantum annealer is limited to a small scale. In this study, we employ the alternating direction method of multipliers. This approach allows us to deal with various types using constraints in the current quantum annealer without slack variables. To test the performance of our algorithm, we use quadratic knapsack problems (QKPs). We compared the accuracy obtained by our method with a simulated annealer and the optimization and sampling mode of a D-Wave machine. As a result of our experiments, we found that the sampling mode shows the best accuracy. We also found that the computational time of our method is faster than that of the exact solver when we tackle various QKPs defined on dense graphs. △ Less

Submitted 10 December, 2020; originally announced December 2020.

Exact solution of free entropy for matrix-valued geometric Brownian motion with non-commutative matrices via the replica method

Authors: Manaka Okuyama, Masayuki Ohzeki

Abstract: Geometric Brownian motion (GBM) is a standard model in stochastic differential equations. In this study, we consider a matrix-valued GBM with non-commutative matrices. Introduction of non-commutative matrices into the matrix-valued GBM makes it difficult to obtain an exact solution because the existence of noise terms prevents diagonalization. However, we show that the replica method enables us to… ▽ More Geometric Brownian motion (GBM) is a standard model in stochastic differential equations. In this study, we consider a matrix-valued GBM with non-commutative matrices. Introduction of non-commutative matrices into the matrix-valued GBM makes it difficult to obtain an exact solution because the existence of noise terms prevents diagonalization. However, we show that the replica method enables us to overcome this difficulty. We map the trace of the time evolution operator of the matrix-valued GBM with non-commutative matrices into the partition function of the isotropic Lipkin-Meshkov-Glick model used in quantum spin systems. Then, solving the eigenvalue problem of the isotropic Lipkin-Meshkov-Glick model, we obtain an analytical expression of the free entropy. Numerical simulation is consistent with our analytical result. Thus, our expression is the exact solution of the free entropy for the matrix-valued GBM with non-commutative matrices. △ Less

Submitted 6 September, 2023; v1 submitted 30 May, 2020; originally announced June 2020.

Comments: 9 pages, 4 figures

Journal ref: J. Phys. Soc. Jpn. 92, 114001 (2023)

Upper bound on the second derivative of the quenched pressure in spin-glass models: weak Griffiths second inequality

Authors: Manaka Okuyama, Masayuki Ohzeki

Abstract: The Griffiths first and second inequalities have played an important role in the analysis of ferromagnetic models. In spin-glass models, although the counterpart of the Griffiths first inequality has been obtained, the counterpart of the Griffiths second inequality has not been established. In this study, we generalize the method in the previous work [J. Phys. Soc. Jpn. 76, 074711 (2007)] to the c… ▽ More The Griffiths first and second inequalities have played an important role in the analysis of ferromagnetic models. In spin-glass models, although the counterpart of the Griffiths first inequality has been obtained, the counterpart of the Griffiths second inequality has not been established. In this study, we generalize the method in the previous work [J. Phys. Soc. Jpn. 76, 074711 (2007)] to the case with multi variables for both symmetric and non-symmetric distributions of the interactions, and derive some correlation inequalities for spin-glass models. Furthermore, by combining the acquired equalities in symmetric distributions, we show that there is a non-trivial positive upper bound on the second derivative of the quenched pressure with respect to the strength of the randomness, which is a weak result of the counterpart of the Griffiths second inequality in spin-glass models for general symmetric distributions. △ Less

Submitted 14 May, 2020; originally announced May 2020.

Comments: 13 pages, 1 figure

Mean field analysis of reverse annealing for code-division multiple-access multiuser detection

Authors: Shunta Arai, Masayuki Ohzeki, Kazuyuki Tanaka

Abstract: We evaluate the typical ARA performance of the CDMA multiuser detection by means of statistical mechanics using the replica method. At first, we consider the oracle cases where the initial candidate solution is randomly generated with a fixed fraction of the original signal in the initial state. In the oracle cases, the first-order phase transition can be avoided or mitigated by ARA if we prepare… ▽ More We evaluate the typical ARA performance of the CDMA multiuser detection by means of statistical mechanics using the replica method. At first, we consider the oracle cases where the initial candidate solution is randomly generated with a fixed fraction of the original signal in the initial state. In the oracle cases, the first-order phase transition can be avoided or mitigated by ARA if we prepare for the proper initial candidate solution. We validate our theoretical analysis with quantum Monte Carlo simulations. The theoretical results to avoid the first-order phase transition are consistent with the numerical results. Next, we consider the practical cases where we prepare for the initial candidate solution obtained by commonly used algorithms. We show that the practical algorithms can exceed the threshold to avoid the first-order phase transition. Finally, we test the performance of ARA with the initial candidate solution obtained by the practical algorithm. In this case, the ARA can not avoid the first-order phase transition even if the initial candidate solution exceeds the threshold to avoid the first-order phase transition. △ Less

Submitted 19 April, 2021; v1 submitted 23 April, 2020; originally announced April 2020.

Comments: 14 pages, 12 figures

Journal ref: Phys. Rev. Research 3, 033006 (2021)

Some inequalities for correlation functions of Ising models with quenched randomness

Authors: Manaka Okuyama, Masayuki Ohzeki

Abstract: Correlation inequalities have played an essential role in the analysis of ferromagnetic models but have not been established in spin glass models. In this study, we obtain some correlation inequalities for the Ising models with quenched randomness, where the distribution of the interactions is symmetric. The acquired inequalities can be regarded as an extension of the previous results, which were… ▽ More Correlation inequalities have played an essential role in the analysis of ferromagnetic models but have not been established in spin glass models. In this study, we obtain some correlation inequalities for the Ising models with quenched randomness, where the distribution of the interactions is symmetric. The acquired inequalities can be regarded as an extension of the previous results, which were limited to the local energy for a spin set, to the local energy for a pair of spin sets. Besides, we also obtain some correlation inequalities for asymmetric distribution. △ Less

Submitted 13 April, 2020; originally announced April 2020.

Comments: 11 pages, 0 figure

Breaking limitation of quantum annealer in solving optimization problems under constraints

Authors: Masayuki Ohzeki

Abstract: Quantum annealing is a generic solver for optimization problems that uses fictitious quantum fluctuation. The most groundbreaking progress in the research field of quantum annealing is its hardware implementation, i.e., the so-called quantum annealer, using artificial spins. However, the connectivity between the artificial spins is sparse and limited on a special network known as the chimera graph… ▽ More Quantum annealing is a generic solver for optimization problems that uses fictitious quantum fluctuation. The most groundbreaking progress in the research field of quantum annealing is its hardware implementation, i.e., the so-called quantum annealer, using artificial spins. However, the connectivity between the artificial spins is sparse and limited on a special network known as the chimera graph. Several embedding techniques have been proposed, but the number of logical spins, which represents the optimization problems to be solved, is drastically reduced. In particular, an optimization problem including fully or even partly connected spins suffers from low embeddable size on the chimera graph. In the present study, we propose an alternative approach to solve a large-scale optimization problem on the chimera graph via a well-known method in statistical mechanics called the Hubbard-Stratonovich transformation or its variants. The proposed method can be used to deal with a fully connected Ising model without embedding on the chimera graph and leads to nontrivial results of the optimization problem. We tested the proposed method with a number of partition problems involving solving linear equations and the traffic flow optimization problem in Sendai and Kyoto cities in Japan. △ Less

Submitted 12 February, 2020; originally announced February 2020.

Comments: 16pages 10 figures

Probing the Universality of Topological Defect Formation in a Quantum Annealer: Kibble-Zurek Mechanism and Beyond

Authors: Yuki Bando, Yuki Susa, Hiroki Oshiyama, Naokazu Shibata, Masayuki Ohzeki, Fernando Javier Gómez-Ruiz, Daniel A. Lidar, Adolfo del Campo, Sei Suzuki, Hidetoshi Nishimori

Abstract: The number of topological defects created in a system driven through a quantum phase transition exhibits a power-law scaling with the driving time. This universal scaling law is the key prediction of the Kibble-Zurek mechanism (KZM), and testing it using a hardware-based quantum simulator is a coveted goal of quantum information science. Here we provide such a test using quantum annealing. Specifi… ▽ More The number of topological defects created in a system driven through a quantum phase transition exhibits a power-law scaling with the driving time. This universal scaling law is the key prediction of the Kibble-Zurek mechanism (KZM), and testing it using a hardware-based quantum simulator is a coveted goal of quantum information science. Here we provide such a test using quantum annealing. Specifically, we report on extensive experimental tests of topological defect formation via the one-dimensional transverse-field Ising model on two different D-Wave quantum annealing devices. We find that the quantum simulator results can indeed be explained by the KZM for open-system quantum dynamics with phase-flip errors, with certain quantitative deviations from the theory likely caused by factors such as random control errors and transient effects. In addition, we probe physics beyond the KZM by identifying signatures of universality in the distribution and cumulants of the number of kinks and their decay, and again find agreement with the quantum simulator results. This implies that the theoretical predictions of the generalized KZM theory, which assumes isolation from the environment, applies beyond its original scope to an open system. We support this result by extensive numerical computations. To check whether an alternative, classical interpretation of these results is possible, we used the spin-vector Monte Carlo model, a candidate classical description of the D-Wave device. We find that the degree of agreement with the experimental data from the D-Wave annealing devices is better for the KZM, a quantum theory, than for the classical spin-vector Monte Carlo model, thus favoring a quantum description of the device. Our work provides an experimental test of quantum critical dynamics in an open quantum system, and paves the way to new directions in quantum simulation experiments. △ Less

Submitted 26 May, 2020; v1 submitted 30 January, 2020; originally announced January 2020.

Report number: RIKEN-iTHEMS-Report-20

Journal ref: Phys. Rev. Research 2, 033369 (2020)

Inequality for local energy of Ising models with quenched randomness and its application

Authors: Manaka Okuyama, Masayuki Ohzeki

Abstract: In this study, we extend the lower bound on the average of the local energy of the Ising model with quenched randomness [J. Phys. Soc. Jpn. 76, 074711 (2007)] obtained for a symmetric distribution to an asymmetric one. Compared with the case of symmetric distribution, our bound has a non-trivial term. By applying the acquired bound to a Gaussian distribution, we obtain the lower bounds on the expe… ▽ More In this study, we extend the lower bound on the average of the local energy of the Ising model with quenched randomness [J. Phys. Soc. Jpn. 76, 074711 (2007)] obtained for a symmetric distribution to an asymmetric one. Compared with the case of symmetric distribution, our bound has a non-trivial term. By applying the acquired bound to a Gaussian distribution, we obtain the lower bounds on the expectation of the square of the correlation function. Thus, we demonstrate that in the Ising model in a Gaussian random field, the spin-glass order parameter generally has a finite value at any temperature, regardless of the forms of the other interactions. △ Less

Submitted 10 April, 2020; v1 submitted 29 January, 2020; originally announced January 2020.

Comments: 6 pages, 0 figure, to appear in J. Phys. Soc. Jpn

Journal ref: J. Phys. Soc. Jpn. 89, 064704 (2020)

A simple relation between frustration and transition points in diluted spin glasses

Authors: Ryoji Miyazaki, Yuta Kudo, Masayuki Ohzeki, Kazuyuki Tanaka

Abstract: We investigate a possible relation between frustration and phase-transition points in spin glasses. The relation is represented as a condition of the number of frustrated plaquettes in the lattice at phase-transition points at zero temperature and was reported to provide very close points to the phase-transition points for several lattices. Although there has been no proof of the relation, the goo… ▽ More We investigate a possible relation between frustration and phase-transition points in spin glasses. The relation is represented as a condition of the number of frustrated plaquettes in the lattice at phase-transition points at zero temperature and was reported to provide very close points to the phase-transition points for several lattices. Although there has been no proof of the relation, the good correspondence in several lattices suggests the validity of the relation and some important role of frustration in the phase transitions. To examine the relation further, we present a natural extension of the relation to diluted lattices and verify its effectiveness for bond-diluted square lattices. We then confirm that the resulting points are in good agreement with the phase-transition points in a wide range of dilution rate. Our result supports the suggestion from the previous work for non-diluted lattices on the importance of frustration to the phase transition of spin glasses. △ Less

Submitted 12 January, 2020; originally announced January 2020.

Comments: 6 pages, 4 figures

Journal ref: Phys. Rev. E 102, 012135 (2020)

Fair Sampling by Simulated Annealing on Quantum Annealer

Authors: Masayuki Yamamoto, Masayuki Ohzeki, Kazuyuki Tanaka

Abstract: Conventional quantum annealing does not sample all ground states fairly. We demonstrate that fair sampling can be achieved by performing simulated annealing on a quantum annealer. We discuss the problems that occur when implementing this method and propose an alternative way to overcome them. We numerically verify the fair sampling ability of our method in a small-scale toy model. Conventional quantum annealing does not sample all ground states fairly. We demonstrate that fair sampling can be achieved by performing simulated annealing on a quantum annealer. We discuss the problems that occur when implementing this method and propose an alternative way to overcome them. We numerically verify the fair sampling ability of our method in a small-scale toy model. △ Less

Submitted 23 December, 2019; originally announced December 2019.

Comments: 2 pages, 2 figures. Submitted to J. Phys. Soc. Jpn

Journal ref: J. Phys. Soc. Jpn. 89, 025002 (2020)

Sparse Modeling in Quantum Many-Body Problems

Authors: Junya Otsuki, Masayuki Ohzeki, Hiroshi Shinaoka, Kazuyoshi Yoshimi

Abstract: This review paper describes the basic concept and technical details of sparse modeling and its applications to quantum many-body problems. Sparse modeling refers to methodologies for finding a small number of relevant parameters that well explain a given dataset. This concept reminds us physics, where the goal is to find a small number of physical laws that are hidden behind complicated phenomena.… ▽ More This review paper describes the basic concept and technical details of sparse modeling and its applications to quantum many-body problems. Sparse modeling refers to methodologies for finding a small number of relevant parameters that well explain a given dataset. This concept reminds us physics, where the goal is to find a small number of physical laws that are hidden behind complicated phenomena. Sparse modeling extends the target of physics from natural phenomena to data, and may be interpreted as "physics for data". The first half of this review introduces sparse modeling for physicists. It is assumed that readers have physics background but no expertise in data science. The second half reviews applications. Matsubara Green's function, which plays a central role in descriptions of correlated systems, has been found to be sparse, meaning that it contains little information. This leads to (i) a new method for solving the ill-conditioned inverse problem for analytical continuation, and (ii) a highly compact representation of Matsubara Green's function, which enables efficient calculations for quantum many-body systems. △ Less

Submitted 11 November, 2019; originally announced November 2019.

Comments: 24 pages, 32 figures, to be published in J. Phys. Soc. Jpn. as an invited review paper

Journal ref: J. Phys. Soc. Jpn. 89, 012001 (2020)

SpM: Sparse modeling tool for analytic continuation of imaginary-time Green's function

Authors: Kazuyoshi Yoshimi, Junya Otsuki, Yuichi Motoyama, Masayuki Ohzeki, Hiroshi Shinaoka

Abstract: We present SpM, a sparse modeling tool for the analytic continuation of imaginary-time Green's function, licensed under GNU General Public License version 3. In quantum Monte Carlo simulation, dynamic physical quantities such as single-particle and magnetic excitation spectra can be obtained by applying analytic continuation to imaginary-time data. However, analytic continuation is an ill-conditio… ▽ More We present SpM, a sparse modeling tool for the analytic continuation of imaginary-time Green's function, licensed under GNU General Public License version 3. In quantum Monte Carlo simulation, dynamic physical quantities such as single-particle and magnetic excitation spectra can be obtained by applying analytic continuation to imaginary-time data. However, analytic continuation is an ill-conditioned inverse problem and thus sensitive to noise and statistical errors. SpM provides stable analytic continuation against noise by means of a modern regularization technique, which automatically selects bases that contain relevant information unaffected by noise. This paper details the use of this program and shows some applications. △ Less

Submitted 5 April, 2019; originally announced April 2019.

Comments: 7 pages, 4 figures

Message-passing algorithm of quantum annealing with nonstoquastic Hamiltonian

Authors: Masayuki Ohzeki

Abstract: Quantum annealing (QA) is a generic method for solving optimization problems using fictitious quantum fluctuation. The current device performing QA involves controlling the transverse field; it is classically simulatable by using the standard technique for mapping the quantum spin systems to the classical ones. In this sense, the current system for QA is not powerful despite utilizing quantum fluc… ▽ More Quantum annealing (QA) is a generic method for solving optimization problems using fictitious quantum fluctuation. The current device performing QA involves controlling the transverse field; it is classically simulatable by using the standard technique for mapping the quantum spin systems to the classical ones. In this sense, the current system for QA is not powerful despite utilizing quantum fluctuation. Hence, we developed a system with a time-dependent Hamiltonian consisting of a combination of the formulated Ising model and the "driver" Hamiltonian with only quantum fluctuation. In the previous study, for a fully connected spin model, quantum fluctuation can be addressed in a relatively simple way. We proved that the fully connected antiferromagnetic interaction can be transformed into a fluctuating transverse field and is thus classically simulatable at sufficiently low temperatures. Using the fluctuating transverse field, we established several ways to simulate part of the nonstoquastic Hamiltonian on classical computers. We formulated a message-passing algorithm in the present study. This algorithm is capable of assessing the performance of QA with part of the nonstoquastic Hamiltonian having a large number of spins. In other words, we developed a different approach for simulating the nonstoquastic Hamiltonian without using the quantum Monte Carlo technique. Our results were validated by comparison to the results obtained by the replica method. △ Less

Submitted 23 January, 2019; v1 submitted 21 January, 2019; originally announced January 2019.

Comments: 17 pages

Control of automated guided vehicles without collision by quantum annealer and digital devices

Authors: Masayuki Ohzeki, Akira Miki, Masamichi J. Miyama, Masayoshi Terabe

Abstract: We formulate an optimization problem to control a large number of automated guided vehicles in a plant without collision. The formulation consists of binary variables. A quadratic cost function over these variables enables us to utilize certain solvers on digital computers and recently developed purpose-specific hardware such as D-Wave 2000Q and the Fujitsu digital annealer. In the present study,… ▽ More We formulate an optimization problem to control a large number of automated guided vehicles in a plant without collision. The formulation consists of binary variables. A quadratic cost function over these variables enables us to utilize certain solvers on digital computers and recently developed purpose-specific hardware such as D-Wave 2000Q and the Fujitsu digital annealer. In the present study, we consider an actual plant in Japan, in which vehicles run, and assess efficiency of our formulation for optimizing the vehicles via several solvers. We confirm that our formulation can be a powerful approach for performing smooth control while avoiding collisions between vehicles, as compared to a conventional method. In addition, comparative experiments performed using several solvers reveal that D-Wave 2000Q can be useful as a rapid solver for generating a plan for controlling the vehicles in a short time although it deals only with a small number of vehicles, while a digital computer can rapidly solve the corresponding optimization problem even with a large number of binary variables. △ Less

Submitted 27 December, 2018; v1 submitted 4 December, 2018; originally announced December 2018.

Comments: 12 pages, 4 figures, some typos are fixed

Dynamics of Order Parameters of Non-stoquastic Hamiltonians in the Adaptive Quantum Monte Carlo Method

Authors: Shunta Arai, Masayuki Ohzeki, Kazuyuki Tanaka

Abstract: We derive macroscopically deterministic flow equations with regard to the order parameters of the ferromagnetic $p$-spin model with infinite-range interactions. The $p$-spin model has a first-order phase transition for $p>2$. In the case of $p\geq5$ ,the $p$-spin model with anti-ferromagnetic XX interaction has a second-order phase transition in a certain region. In this case, however, the model b… ▽ More We derive macroscopically deterministic flow equations with regard to the order parameters of the ferromagnetic $p$-spin model with infinite-range interactions. The $p$-spin model has a first-order phase transition for $p>2$. In the case of $p\geq5$ ,the $p$-spin model with anti-ferromagnetic XX interaction has a second-order phase transition in a certain region. In this case, however, the model becomes a non-stoqustic Hamiltonian, resulting in a negative sign problem. To simulate the $p$-spin model with anti-ferromagnetic XX interaction, we utilize the adaptive quantum Monte Carlo method. By using this method, we can regard the effect of the anti-ferromagnetic XX interaction as fluctuations of the transverse magnetic field. A previous study derived deterministic flow equations of the order parameters in the quantum Monte Carlo method. In this study, we derive macroscopically deterministic flow equations for the magnetization and transverse magnetization from the master equation in the adaptive quantum Monte Carlo method. Under the Suzuki-Trotter decomposition, we consider the Glauber-type stochastic process. We solve these differential equations by using the Runge-Kutta method and verify that these results are consistent with the saddle-point solution of mean-field theory. Finally, we analyze the stability of the equilibrium solutions obtained by the differential equations. △ Less

Submitted 12 February, 2019; v1 submitted 23 October, 2018; originally announced October 2018.

Comments: 10 pages, 9 figures

Journal ref: Phys. Rev. E 99, 032120 (2019)

Steady state distributions of network of degenerate optical parametric oscillators in solving combinatorial optimization problems

Authors: Ryoji Miyazaki, Masayuki Ohzeki

Abstract: We investigate network of degenerate optical parametric oscillators (DOPOs) as a model of the coherent Ising machine, an architecture for solving Ising problems. The network represents the interaction in the Ising model, which is a generalization of a previously proposed one for the two-DOPO case. Dynamics of the DOPOs is described by the Fokker-Planck equation in the positive $P$ representation.… ▽ More We investigate network of degenerate optical parametric oscillators (DOPOs) as a model of the coherent Ising machine, an architecture for solving Ising problems. The network represents the interaction in the Ising model, which is a generalization of a previously proposed one for the two-DOPO case. Dynamics of the DOPOs is described by the Fokker-Planck equation in the positive $P$ representation. We obtain approximate steady state distributions for arbitrary Ising problems under some ansatz. Using the method of statistical mechanics, we analytically demonstrate that the most probable states in a particular range of the parameters correspond to the true optimal states for two rather simple problems, i.e., fully-connected ferromagnetic coupling without/with binary random fields. In particular, for the random-field problem, the distribution correctly detects the phase transition that occurs in the target Ising model with varying the magnitude of the fields. △ Less

Submitted 31 August, 2018; originally announced August 2018.

Comments: 13 pages, 2 figures

Journal ref: Phys. Rev. A 98, 053839 (2018)

An exact solution of the partition function for mean-field quantum spin systems without the static approximation

Authors: Manaka Okuyama, Masayuki Ohzeki

Abstract: Suzuki-Trotter decomposition is a well-known technique used to calculate the partition function of quantum spin systems, in which the imaginary-time dependence of the partition function occurs inevitably. Since it is very difficult to explicitly treat the imaginary-time dependence of the partition function, we usually neglect the imaginary-time dynamical effect, which is called the static approxim… ▽ More Suzuki-Trotter decomposition is a well-known technique used to calculate the partition function of quantum spin systems, in which the imaginary-time dependence of the partition function occurs inevitably. Since it is very difficult to explicitly treat the imaginary-time dependence of the partition function, we usually neglect the imaginary-time dynamical effect, which is called the static approximation. Although the static approximation is the first approach, it is not even clear when the static approximation is justified for mean-field quantum spin systems, that is, mean-field quantum spin systems have not been solved exactly so far. In this study, we solve exactly the partition function for a particular class of mean-field quantum spin systems including randomness without the static approximation. The partition function can be regarded as a result of time evolution in the imaginary-time Schrödinger equation, and solving the exact solution of the partition function is equivalent to solving the optimal control problem in the imaginary-time Schrödinger equation. As the result, the solution of the optimal control problem coincides exactly with the static approximate solution of the partition function and, therefore, the static approximation is exact for the particular class of mean-field quantum spin systems including randomness in general. Furthermore, we prove that the analysis of the previous study in quantum annealing is exact where the non-stoquastic interaction and the inhomogeneous transverse field accelerate the computational time exponentially for mean-field quantum spin systems. △ Less

Submitted 29 August, 2018; originally announced August 2018.

Comments: 13 pages, 0 figure

Optimization of neural networks via finite-value quantum fluctuations

Authors: Masayuki Ohzeki, Shuntaro Okada, Masayoshi Terabe, Shinichiro Taguchi

Abstract: We numerically test an optimization method for deep neural networks (DNNs) using quantum fluctuations inspired by quantum annealing. For efficient optimization, our method utilizes the quantum tunneling effect beyond the potential barriers. The path integral formulation of the DNN optimization generates an attracting force to simulate the quantum tunneling effect. In the standard quantum annealing… ▽ More We numerically test an optimization method for deep neural networks (DNNs) using quantum fluctuations inspired by quantum annealing. For efficient optimization, our method utilizes the quantum tunneling effect beyond the potential barriers. The path integral formulation of the DNN optimization generates an attracting force to simulate the quantum tunneling effect. In the standard quantum annealing method, the quantum fluctuations will vanish at the last stage of optimization. In this study, we propose a learning protocol that utilizes a finite value for quantum fluctuations strength to obtain higher generalization performance, which is a type of robustness. We demonstrate the performance of our method using two well-known open datasets: the MNIST dataset and the Olivetti face dataset. Although computational costs prevent us from testing our method on large datasets with high-dimensional data, results show that our method can enhance generalization performance by induction of the finite value for quantum fluctuations. △ Less

Submitted 1 July, 2018; originally announced July 2018.

Comments: 11 pages, 3 figures

A useful fundamental speed limit for the imaginary-time Schrodinger equation

Authors: Manaka Okuyama, Masayuki Ohzeki

Abstract: The quantum speed limit (QSL), or the energy-time uncertainty relation, gives a fundamental speed limit for quantum dynamics. Recently, Kieu [arXiv:1702.00603] derived a new class of QSL which is not only formal but also suitable for actually evaluating the speed limit. Inspired by his work, we obtain a similar speed limit for the imaginary-time Schrödinger equation. Using this new bound, we show… ▽ More The quantum speed limit (QSL), or the energy-time uncertainty relation, gives a fundamental speed limit for quantum dynamics. Recently, Kieu [arXiv:1702.00603] derived a new class of QSL which is not only formal but also suitable for actually evaluating the speed limit. Inspired by his work, we obtain a similar speed limit for the imaginary-time Schrödinger equation. Using this new bound, we show that the optimal computational time of the Grover problem in imaginary-time quantum annealing is bounded from below by $\log N$, which is consistent with a result of previous study. △ Less

Submitted 23 June, 2018; originally announced June 2018.

Comments: 4 pages, 0 figure

Momentum-Space Renormalization Group Transformation in Bayesian Image Modeling by Gaussian Graphical Model

Authors: Kazuyuki Tanaka, Masamichi Nakamura, Shun Kataoka, Masayuki Ohzeki, Muneki Yasuda

Abstract: A new Bayesian modeling method is proposed by combining the maximization of the marginal likelihood with a momentum-space renormalization group transformation for Gaussian graphical models. Moreover, we present a scheme for computint the statistical averages of hyperparameters and mean square errors in our proposed method based on a momentumspace renormalization transformation. A new Bayesian modeling method is proposed by combining the maximization of the marginal likelihood with a momentum-space renormalization group transformation for Gaussian graphical models. Moreover, we present a scheme for computint the statistical averages of hyperparameters and mean square errors in our proposed method based on a momentumspace renormalization transformation. △ Less

Submitted 19 March, 2018; originally announced April 2018.

Comments: 6 pages, 1 figure

Statistical-mechanical analysis of compressed sensing for Hamiltonian estimation of Ising spin glass

Authors: Chako Takahashi, Masayuki Ohzeki, Shuntaro Okada, Masayoshi Terabe, Shinichiro Taguchi, Kazuyuki Tanaka

Abstract: Several powerful machines, such as the D-Wave 2000Q, dedicated to solving combinatorial optimization problems through the Ising-model formulation have been developed. To input problems into the machines, the unknown parameters on the Ising model must be determined, and this is necessarily a nontrivial task. It could be beneficial to construct a method to estimate the parameters of the Ising model… ▽ More Several powerful machines, such as the D-Wave 2000Q, dedicated to solving combinatorial optimization problems through the Ising-model formulation have been developed. To input problems into the machines, the unknown parameters on the Ising model must be determined, and this is necessarily a nontrivial task. It could be beneficial to construct a method to estimate the parameters of the Ising model from several pairs of values of the energy and spin configurations. In the present paper, we propose a simple method employing the $L_1$-norm minimization, which is based on the concept of the compressed sensing. Moreover, we analyze the typical performance of our proposed method of the Hamiltonian estimation by using the replica method. We also compare our analytical results through several numerical experiments using the alternating direction method of multipliers. △ Less

Submitted 6 March, 2018; originally announced March 2018.

Comments: 8 pages, 2 figures

Overcomplete compact representation of two-particle Green's functions

Authors: Hiroshi Shinaoka, Junya Otsuki, Kristjan Haule, Markus Wallerberger, Emanuel Gull, Kazuyoshi Yoshimi, Masayuki Ohzeki

Abstract: Two-particle Green's functions and the vertex functions play a critical role in theoretical frameworks for describing strongly correlated electron systems. However, numerical calculations at two-particle level often suffer from large computation time and massive memory consumption. We derive a general expansion formula for the two-particle Green's functions in terms of an overcomplete representati… ▽ More Two-particle Green's functions and the vertex functions play a critical role in theoretical frameworks for describing strongly correlated electron systems. However, numerical calculations at two-particle level often suffer from large computation time and massive memory consumption. We derive a general expansion formula for the two-particle Green's functions in terms of an overcomplete representation based on the recently proposed "intermediate representation" basis. The expansion formula is obtained by decomposing the spectral representation of the two-particle Green's function. We demonstrate that the expansion coefficients decay exponentially, while all high-frequency and long-tail structures in the Matsubara-frequency domain are retained. This representation therefore enables efficient treatment of two-particle quantities and opens a route to the application of modern many-body theories to realistic strongly correlated electron systems. △ Less

Submitted 9 May, 2018; v1 submitted 5 March, 2018; originally announced March 2018.

Comments: 17 pages, 12 figures (minor modifications)

Journal ref: Phys. Rev. B 97, 205111 (2018)

Deep Neural Network Detects Quantum Phase Transition

Authors: Shunta Arai, Masayuki Ohzeki, Kazuyuki Tanaka

Abstract: We detect the quantum phase transition of a quantum many-body system by mapping the observed results of the quantum state onto a neural network. In the present study, we utilized the simplest case of a quantum many-body system, namely a one-dimensional chain of Ising spins with the transverse Ising model. We prepared several spin configurations, which were obtained using repeated observations of t… ▽ More We detect the quantum phase transition of a quantum many-body system by mapping the observed results of the quantum state onto a neural network. In the present study, we utilized the simplest case of a quantum many-body system, namely a one-dimensional chain of Ising spins with the transverse Ising model. We prepared several spin configurations, which were obtained using repeated observations of the model for a particular strength of the transverse field, as input data for the neural network. Although the proposed method can be employed using experimental observations of quantum many-body systems, we tested our technique with spin configurations generated by a quantum Monte Carlo simulation without initial relaxation. The neural network successfully classified the strength of transverse field only from the spin configurations, leading to consistent estimations of the critical point of our model $Γ_c =J$. △ Less

Submitted 1 December, 2017; originally announced December 2017.

Comments: 4pages,3 figures

Journal ref: J. Phys. Soc. Jpn., Vol.87, No.3,2018

An extension of estimation of critical points in ground state for random spin systems

Authors: Masayuki Ohzeki, Yuta Kudo, Kazuyuki Tanaka

Abstract: Most of the analytical studies on spin glasses are performed by using mean-field theory and renormalization group analysis. Analytical studies on finite-dimensional spin glasses are very challenging. In this short note, a possible exten- sion of the approaches on the phase transition in spin glasses is demonstrated. To validate our extension, we compared our estimates on the critical points with t… ▽ More Most of the analytical studies on spin glasses are performed by using mean-field theory and renormalization group analysis. Analytical studies on finite-dimensional spin glasses are very challenging. In this short note, a possible exten- sion of the approaches on the phase transition in spin glasses is demonstrated. To validate our extension, we compared our estimates on the critical points with the existing numerical results. △ Less

Submitted 9 November, 2017; originally announced November 2017.

Comments: 2 pages, 2 figures