Noboru Murata

• Dr. Eng.
• Professor (Full) at Waseda University

This paper presents a novel topology optimization approach for the design of synchronous reluctance motors based on an autoencoder (AE) combined with the level set (LS) method. As the initial shapes of the LS method, the technique uses the shape generated by the AE, which learns the relationship between the objective function values and the design...

Despite the importance of sleep to the cerebral cortex, how much sleep changes cortical neuronal firing remains unclear due to complicated firing behaviors. Here we quantified firing of cortical neurons using Hawkes process modeling that can model sequential random events exhibiting temporal clusters. "Intensity" is a parameter of Hawkes process th...

The Expectation–Maximization (EM) algorithm is a simple meta-algorithm that has been used for many years as a methodology for statistical inference when there are missing measurements in the observed data or when the data is composed of observables and unobservables. Its general properties are well studied, and also, there are countless ways to app...

The Expectation--Maximization (EM) algorithm is a simple meta-algorithm that has been used for many years as a methodology for statistical inference when there are missing measurements in the observed data or when the data is composed of observables and unobservables. Its general properties are well studied, and also, there are countless ways to ap...

Purpose The density method is one of the powerful topology optimization methods of magnetic devices. The density method has the advantage that it has a high degree of freedom of shape expression which results in a high-performance design. On the other hand, it has also the drawback that unsuitable shapes for actually manufacturing are likely to be...

A large number of neurons form cell assemblies that process information in the brain. Recent developments in measurement technology, one of which is calcium imaging, have made it possible to study cell assemblies. In this study, we aim to extract cell assemblies from calcium imaging data. We propose a clustering approach based on non-negative matri...

We proposed a new multi-objective optimization approach combining both gradient calculation in actual high-dimensional design space and that in a low-dimensional latent space that is an appropriately compressed design space with Auto-Encoder. The proposed approach enables us to quickly and precisely search for more global pareto solution group in m...

Single-molecule localization microscopy is a widely used technique in biological research for measuring the nanostructures of samples smaller than the diffraction limit. This study uses multifocal plane microscopy and addresses the three-dimensional (3D) single-molecule localization problem, where lateral and axial locations of molecules are estima...

Single-molecule localization microscopy is widely used in biological research for measuring the nanostructures of samples smaller than the diffraction limit. In this paper, a novel method for regression of the coordinates of molecules for multifocal plane microscopy is presented. A regression problem for the target space is decomposed into regressi...

Single molecule localization microscopy is widely used in biological research for measuring the nanostructures of samples smaller than the diffraction limit. This study uses multifocal plane microscopy and addresses the 3D single molecule localization problem, where lateral and axial locations of molecules are estimated. However, when we multifocal...

In this paper, we examine a geometrical projection algorithm for statistical inference. The algorithm is based on Pythagorean relation and it is derivative-free as well as representation-free that is useful in nonparametric cases. We derive a bound of learning rate to guarantee local convergence. In special cases of m-mixture and e-mixture estimati...

In this paper, we examine a geometrical projection algorithm for statistical inference. The algorithm is based on Pythagorean relation and it is derivative-free as well as representation-free that is useful in nonparametric cases. We derive a bound of learning rate to guarantee local convergence. In special cases of m-mixture and e-mixture estimati...

Sleep is an essential process for the survival of animals. However, its phenomenon is poorly understood. To understand the phenomenon of sleep, the analysis should be made from the activities of a large number of cortical neurons. Calcium imaging is a recently developed technique that can record a large number of neurons simultaneously, however, it...

Outlier detection is used to identify data points or a small number of subsets of data that are significantly different from most other data in a given dataset. It is challenging to detect outliers using an objective and quantitative approach. Methods that use the framework of statistical hypothesis testing are widely used by assuming a specific pa...

Modal linear regression (MLR) is used for modeling the conditional mode of a response as a linear predictor of explanatory variables. It is an effective approach to dealing with response variables having a multimodal distribution or those contaminated by outliers. Because of the semiparametric nature of MLR, constructing a statistical model manifol...

This study focuses on darknet traffic analysis and applies tensor factorization in order to detect coordinated group activities, such as a botnet. Tensor factorization is a powerful tool for extracting co-occurrence patterns that is highly interpretable and can handle more variables than matrix factorization. We propose a simple method for detectin...

To maintain acceptable levels of security, organizations must manage their IT assets and related vulnerabilities. However, this can be a considerable burden because their resources are often limited. We have been working on a technique and system architecture that monitor the vulnerability of the IT assets on an organization's administrative networ...

Sparse representation is a signal model to represent signals with a linear combination of a small number of prototype signals called atoms, and a set of atoms is called a dictionary. The design of the dictionary is a fundamental problem for sparse representation. However, when there are scaled or translated features in the signals, unstructured dic...

As the solar photovoltaic systems (PV) are massively installed into power system on an urban scale, short-term (about 10 seconds) fluctuations of PV output can negatively effect on the power system. Battery storage systems are capable of alleviating the effect of PV fluctuations; power flow simulation of such systems is also meaningful to discuss t...

Electroencephalography (EEG) is a non-invasive brain imaging technique that describes neural electrical activation with good temporal resolution. Source localization is required for clinical and functional interpretations of EEG signals, and most commonly is achieved via the dipole model; however, the number of dipoles in the brain should be determ...

We have obtained an integral representation of the shallow neural network that attains the global minimum of its backpropagation (BP) training problem. According to our unpublished numerical simulations conducted several years prior to this study, we had noticed that such an integral representation may exist, but it was not proven until today. Firs...

The introduction of photovoltaic power systems is being significantly promoted. This paper proposes the implementation of a distributed energy management framework linking demand-side management systems and supply-side management system under the given time-of-use pricing program for efficient utilization of photovoltaic power outputs; each system...

The feature map obtained from the denoising autoencoder (DAE) is investigated by determining transportation dynamics of the DAE, which is a cornerstone for deep learning. Despite the rapid development in its application, deep neural networks remain analytically unexplained, because the feature maps are nested and parameters are not faithful. In thi...

The feature map obtained from the denoising autoencoder (DAE) is investigated by determining transportation dynamics of the DAE, which is a cornerstone for deep learning. Despite the rapid development in its application, deep neural networks remain analytically unexplained, because the feature maps are nested and parameters are not faithful. In thi...

Plasticity is one of the most important features of the nervous systems that enable animals to adjust their behavior to an ever-changing external world. A major mechanism of plasticity is the changes in synaptic efficacy between neurons, and therefore estimation of neural connections is crucial for investigating information processing in the brain....

Plasticity is one of the most important properties of the nervous system, which enables animals to adjust their behavior to the ever-changing external environment. Changes in synaptic efficacy between neurons constitute one of the major mechanisms of plasticity. Therefore, estimation of neural connections is crucial for investigating information pr...

The continuum limit is an effective method for modeling complex discrete structures such as deep neural networks to facilitate their interpretability. The continuum limits of deep networks are investigated with respect to two directions: width and depth. The width continuum limit is a limit of the linear combination of functions, or a continuous mo...

We propose a method for intrinsic dimension estimation. By fitting the power of distance from an inspection point and the number of samples included inside a ball with a radius equal to the distance, to a regression model, we estimate the goodness of fit. Then, by using the maximum likelihood method, we estimate the local intrinsic dimension around...

A number of image super resolution algorithms based on the sparse coding have successfully implemented multi-frame super resolution in recent years. In order to utilize multiple low-resolution observations, both accurate image registration and sparse coding are required. Previous study on multi-frame super resolution based on sparse coding firstly...

In a product market or stock market, different products or stocks compete for the same consumers or purchasers. We propose a method to estimate the time-varying transition matrix of the product share using a multivariate time series of the product share. The method is based on the assumption that each of the observed time series of shares is a stat...

Program code and dataset to reproduce the results. Python code for the proposed method, and original dataset are available as a supporting information file. (ZIP)

Unexpected fluctuation of wind power output will become serious problems from the viewpoint of stable supply for an electricity grid. Operating a battery system installed in the grid for mitigating the short-term fluctuation is one of the new approaches for grid stabilization. In this paper, we propose a method of generating synthetic wind power pr...

A method for botnet detection from traffic data of the Internet by the Non-negative Matrix Factorization (NMF) was proposed by (Yamauchi et al. 2012). This method assumes that traffic data is composed by several types of communications, and estimates the number of types in the data by the minimum description length (MDL) criterion. However, conside...

A method for estimating Shannon differential entropy is proposed based on the second order expansion of the probability mass around the inspection point with respect to the distance from the point. Polynomial regression with Poisson error structure is utilized to estimate the values of density function. The density estimates at every given data poi...

Mixture modeling is one of the simplest ways to represent complicated probability density functions, and to integrate information from different sources. There are two typical mixtures in the context of information geometry, the m- and e-mixtures. This paper proposes a novel framework of non-parametric e-mixture modeling by using a simple estimatio...

This study considers the common situation in data analysis when there are few observations of the distribution of interest or the target distribution, while abundant observations are available from auxiliary distributions. In this situation, it is natural to compensate for the lack of data from the target distribution by using data sets from these...

There are a large number of image super resolution algorithms based on the sparse coding, and some algorithms realize multi-frame super resolution. For utilizing multiple low resolution observations, both accurate image registration and sparse coding are required. Previous study on multi-frame super resolution based on sparse coding firstly apply b...

Data representation in a stacked denoising autoencoder is investigated. Decoding is a simple technique for translating a stacked denoising autoencoder into a composition of denoising autoencoders in the ground space. In the infinitesimal limit, a composition of denoising autoencoders is reduced to a continuous denoising autoencoder, which is rich i...

We propose a method to extract connectivity between neurons for extracellularly recorded multiple spike trains. The method removes pseudo-correlation caused by propagation of information along an indirect pathway, and is also robust against the influence from unobserved neurons. The estimation algorithm consists of iterations of a simple matrix inv...

Some of important methods for signal processing, such as principal component analysis (PCA), independent component analysis (ICA), non-negative matrix factorization (NMF), and sparse representation (SR), can be discussed in a unified framework where a data matrix is decomposed into a product of two specific matrices. Differences of those methods ar...

Background: Knowledge about the distribution, strength, and direction of synaptic connections within neuronal networks are crucial for understanding brain function. Electrophysiology using multiple electrodes provides a very high temporal resolution, but does not yield sufficient spatial information for resolving neuronal connection topology. Opti...

A large number of image super resolution algorithms based on the sparse coding are proposed, and some algorithms realize the multi-frame super resolution. In multi-frame super resolution based on the sparse coding, both accurate image registration and sparse coding are required. Previous study on multi-frame super resolution based on sparse coding...

This paper investigates the approximation property of the neural network with unbounded activation functions, such as the rectified linear unit (ReLU), which is new de-facto standard of deep learning. The ReLU network can be analyzed by the ridgelet transform with respect to Lizorkin distributions, which is introduced in this paper. By showing two...

An optimal operational planning problem of residential energy system has been formulated by Mixed Integer Linear Programming (MILP). The decision variables of optimal operational planning problem are energy and mass flows, equipment's operating statuses, and energy level of storage. Many kinds of energy supply equipment are available for householde...

In this paper, the limitation that is prominent in most existing works of change-point detection methods is addressed by proposing a nonparametric, computationally efficient method. The limitation is that most works assume that each data point observed at each time step is a single multi-dimensional vector. However, there are many situations where...

Estimators for differential entropy are proposed. The estimators are based on the second order expansion of the probability mass around the inspection point with respect to the distance from the point. Simple linear regression is utilized to estimate the values of density function and its second derivative at a point. After estimating the values of...

A classification framework using only a set of distance matrices is proposed. The proposed algorithm can learn a classifier only from a set of distance matrices or similarity matrices, hence applicable to structured data, which do not have natural vector representation such as time series and graphs. Random forest is used to explore ideal feature r...

An image super-resolution method from multiple observation of low-resolution images is proposed. The method is based on sub-pixel accuracy block matching for estimating relative displacements of observed images, and sparse signal representation for estimating the corresponding high-resolution image, where correspondence between high- and low-resolu...

A simple temporal point process (SPP) is an important class of time series, where the sample realization of the process is solely composed of the times at which events occur. Particular examples of point process data are neuronal spike patterns or spike trains, and a large number of distance and similarity metrics for those data have been proposed....

Clustering is a fundamental tool for exploratory data analysis. Information theoretic clustering is based on the optimization of information theoretic quantities such as entropy and mutual information. Recently, since these quantities can be estimated in non-parametric manner, non-parametric information theoretic clustering gains much attention. As...

A problem of estimating the intrinsic graph structure from observed data is considered. The observed data in this study is a matrix with elements representing dependency between nodes in the graph. Each element of the observed matrix represents, for example, co-occurrence of events at two nodes, or correlation of variables corresponding to two node...

A new sampling learning method for neural networks is proposed. Derived from an integral representation of neural networks, an oracle probability distribution of hidden parameters is introduced. In general rigorous sampling from the oracle distribution holds numerical difficulty, a linear-time sampling algorithm is also developed. Numerical experim...

The main objectives of this study are to consider both thermal comfort and energy consumption on operational planning problem of residential energy system, and to handle uncertainty of energy demand and PV output for the future as a scenario-based stochastic programming problem. The energy system consists of photovoltaic power generator, electrical...

Forecast of the energy demand is an important topic for the realization of effective energy management. In this paper, the authors focus on the K-nearest neighbor approach for forecast of the energy demand pattern, introduce an idea of distance metric learning to select appropriate K-nearest neighbors, and propose some learning frameworks for forec...

Nowadays, PEFC-CGS is getting attention as a distributed energy system. It has high efficiency, and a lot of studies of PEFC-CGS were reported. But almost of them didn't consider energy prediction error. In this study, in order to evaluate energy-saving performance of PEFC-CGS for residential use with energy prediction error, I made PEFC-CGS contro...

The subject of this study is to plan the operation of residential energy system with uncertain parameters based on ex-ante decision before uncertain parameters are realized. This paper applies a scenario-based stochastic programming framework to the operational planning problem having uncertain energy demand as parameters. Based on predicted energy...

Clustering is a representative of unsupervised learning and one of the important approaches in exploratory data analysis. By its very nature, clustering without strong assumption on data distribution is desirable. Information-theoretic clustering is a class of clustering methods that optimize information-theoretic quantities such as entropy and mut...

A graph is a mathematical representation of a set of variables where some pairs of the variables are connected by edges. Common examples of graphs are railroads, the Internet, and neural networks. It is both theoretically and practically important to estimate the intensity of direct connections between variables. In this study, a problem of estimat...

To realize stable production in the steel industry, it is important to control molten steel temperature in a continuous casting process. The present work aims to provide a general framework of gray-box modeling and to develop a gray-box model that predicts and controls molten steel temperature in a tundish (TD temp) with high accuracy. Since the ad...

An image super-resolution method from multiple observation of low-resolution images is proposed. The method is based on sub-pixel accuracy block matching for estimating relative displacements of observed images, and sparse signal representation for estimating the corresponding high-resolution image. Relative displacements of small patches of observ...

A new initialization method for hidden parameters in a neural network is proposed. Derived from the integral representation of the neural network, a nonparametric probability distribution of hidden parameters is introduced. In this proposal, hidden parameters are initialized by samples drawn from this distribution, and output parameters are fitted...

The importance of dimension reduction has been increasing according to the growth of the size of available data in many fields. An appropriate dimension reduction method of raw data helps to reduce computational time and to expose the intrinsic structure of complex data. Sliced inverse regression is a well-known dimension reduction method for regre...

Sparse signal models have been the focus of recent research. In sparse coding, signals are represented with a linear combination of a small number of elementary signals called atoms, and the collection of atoms is called a dictionary. Design of the dictionary has strong influence on the signal approximation performance. Recently, to put prior infor...

The Shannon information content is a valuable numerical characteristic of probability distributions. The problem of estimating the information content from an observed dataset is very important in the fields of statistics, information theory, and machine learning. The contribution of the present paper is in proposing information estimators, and sho...

When evaluating residential energy systems like co-generation systems, hot water and electricity demand profiles are critical. In this paper, the authors aim to extract basic time-series demand patterns from two kinds of measured demand (electricity and domestic hot water), and also aim to reveal effective demand patterns for primary energy saving....

A novel gray-box model is proposed to estimate molten steel temperature in a continuous casting process at a steel making plant by combining a first-principle model and a statistical model. The first-principle model was developed on the basis of computational fluid dynamics (CFD) simulations to simplify the model and to improve estimation accuracy....

Clustering a given set of data is crucial in many fields including image processing. It plays important roles in image segmentation and object detection for example. This paper proposes a framework of building a similarity matrix for a given dataset, which is then used for clustering the dataset. The similarity between two points are defined based...

This paper considers N − 1-dimensional hypersurface fitting based on L 2 distance in N-dimensional input space. The problem is usually reduced to hyperplane fitting in higher dimension. However, because feature mapping is generally a nonlinear mapping, it does not preserve the order of lengthes, and this derives an unacceptable fitting result. To a...

An appropriate dimension reduction of raw data helps to reduce computational time and to reveal the intrinsic structure of complex data. In this paper, a di-mension reduction method for regression is proposed. The method is based on the well-known sliced inverse regression and conditional entropy minimization. Using entropy as a measure of dispersi...

Electricity consumption in households varies dependent on a lot of possible reasons such as lifestyle, family configuration, and weather. It is of great importance to optimize the electricity generation system to install for each household. In our previous work, we proposed a clustering approach for extracting a small number of basic electricity co...

Controlling temperature of molten steel is crucial for product quality in continuous casting. In this paper, sensitivity analysis is carried out on a statistical model for predicting temperature in tundish, and important and influential operations on temperature are identified.

A statistical model for predicting the liquid steel temperature in the ladle and in the tundish is devel-oped. Given a large data set in a steelmaking process, the proposed model predicts the temperature in a seconds with a good accuracy. The data are divided into four phases at the mediation of five temperature measurements: before tapping from th...

In this paper, generalised statistical independence in statistical models for categorical distributions is proposed from the viewpoint of generalised multiplication characterised by a monotonically increasing function and its inverse function, and it is implemented in naive Bayes models. This paper also proposes an idea of their estimation method w...

Kernel methods are known to be effective for nonlinear multivariate analysis. One of the main issues in the practical use of kernel methods is the selection of kernel. There have been a lot of studies on kernel selection and kernel learning. Multiple kernel learning (MKL) is one of the promising kernel optimization approaches. Kernel methods are ap...

Nonnegative Matrix Factorization (NMF) is broadly used as a mathematical tool for processing tasks of tabulated data. In this paper, an extension of NMF based on a generalized product rule, defined with a nonlinear one-parameter function and its inverse, is proposed. From a viewpoint of subspace methods, the extended NMF constructs flexible subspac...

Sparse coding is an important optimization problem with numerous applications. In this paper, we describe the problem and the commonly used pursuit methods, and propose a best-first tree search algorithm employing multiple queues for unexplored tree nodes. We as-sess the effectiveness of our method in an extensive computational ex-periment, showing...

Renewable energy, for example photovoltaic (PV) and wind electricity, commands attention as a one of the solutions to environmental problems. However, its generating power depends on the meteorological condition. So, forecasting the output of renewable energy is important for efficient energy management. In this research, we focused on PV and carri...

Recently, remarkable developments of new energy technologies have been achieved against various energy prob-lems. Photovoltaic (PV) system, one of such technologies, has an advantage of utilizing infinite and clean energy. On the contrary, it also has a disadvantage of unreliable power supply mainly caused by unstable weather. The fluctuation of th...

Solar power, wind power, and co-generation (com-bined heat and power) systems are possible candidate for household power generation. These systems have their advan-tages and disadvantages. To propose the optimal combination of the power generation systems, the extraction of basic patterns of energy consumption of the house is required. In this stud...

For calibration of general radially symmetric distortion of omnidirectional cameras such as fish-eye lenses, calibration parameters are usually estimated so that curved lines, which are supposed to be straight in the real-world, are mapped to straight lines in the calibrated image, which is called plumbline principle. Under the principle, the camer...

We developed a new speaker verification system that is robust to intra-speaker variation. There is a strong likelihood that intra-speaker variations will occur due to changes in talking styles, the periods when an individual speaks, and so on. It is well known that such variation generally degrades the performance of speaker verification systems. T...

In this paper, two methods for one-dimensional reduction of data by hyperplane fitting are proposed. One is least α-percentile of squares, which is an extension of least median of squares estimation and minimizes the α-percentile of squared Euclidean distance. The other is least k-th power deviation, which is an extension of least squares estimatio...

The Shannon information content is a fundamental quantity and it is of great importance to estimate it from observed dataset in the field of statistics, information theory, and machine learning. In this study, an estimator for the information content using a given set of weighted data is proposed. The empirical data distribution varies depending o...

We applied a multiple kernel learning (MKL) method based on information-theoretic optimization to speaker recognition. Most of the kernel methods applied to speaker recognition systems require a suitable kernel function and its parameters to be determined for a given data set. In contrast, MKL eliminates the need for strict determination of the ker...

The Bradley-Terry model is a statistical representation for one's preference or ranking data by using pairwise comparison results of items. For estimation of the model, several methods based on the sum of weighted Kullback-Leibler divergences have been proposed from various contexts. The purpose of this letter is to interpret an estimation mechanis...

We derive and demonstrate new methods for dewarping images depicted in convex mirrors in artwork and for estimating the three-dimensional shapes of the mirrors themselves. Previous methods were based on the assumption that mirrors were spherical or paraboloidal, an assumption unlikely to hold for hand-blown glass spheres used in early Renaissance a...