Showing 1–14 of 14 results for author: Iwata, M

Multi-pooling 3D Convolutional Neural Network for fMRI Classification of Visual Brain States

Authors: Zhen Zhang, Masaki Takeda, Makoto Iwata

Abstract: Neural decoding of visual object classification via functional magnetic resonance imaging (fMRI) data is challenging and is vital to understand underlying brain mechanisms. This paper proposed a multi-pooling 3D convolutional neural network (MP3DCNN) to improve fMRI classification accuracy. MP3DCNN is mainly composed of a three-layer 3DCNN, where the first and second layers of 3D convolutions each… ▽ More Neural decoding of visual object classification via functional magnetic resonance imaging (fMRI) data is challenging and is vital to understand underlying brain mechanisms. This paper proposed a multi-pooling 3D convolutional neural network (MP3DCNN) to improve fMRI classification accuracy. MP3DCNN is mainly composed of a three-layer 3DCNN, where the first and second layers of 3D convolutions each have a branch of pooling connection. The results showed that this model can improve the classification accuracy for categorical (face vs. object), face sub-categorical (male face vs. female face), and object sub-categorical (natural object vs. artificial object) classifications from 1.684% to 14.918% over the previous study in decoding brain mechanisms. △ Less

Submitted 25 March, 2023; originally announced March 2023.

Report number: 2023 IEEE Conference on Artificial Intelligence (CAI)

Technical Progress Analysis Using a Dynamic Topic Model for Technical Terms to Revise Patent Classification Codes

Authors: Mana Iwata, Yoshiro Matsuda, Yoshimasa Utsumi, Yoshitoshi Tanaka, Kazuhide Nakata

Abstract: Japanese patents are assigned a patent classification code, FI (File Index), that is unique to Japan. FI is a subdivision of the IPC, an international patent classification code, that is related to Japanese technology. FIs are revised to keep up with technological developments. These revisions have already established more than 30,000 new FIs since 2006. However, these revisions require a lot of t… ▽ More Japanese patents are assigned a patent classification code, FI (File Index), that is unique to Japan. FI is a subdivision of the IPC, an international patent classification code, that is related to Japanese technology. FIs are revised to keep up with technological developments. These revisions have already established more than 30,000 new FIs since 2006. However, these revisions require a lot of time and workload. Moreover, these revisions are not automated and are thus inefficient. Therefore, using machine learning to assist in the revision of patent classification codes (FI) will lead to improved accuracy and efficiency. This study analyzes patent documents from this new perspective of assisting in the revision of patent classification codes with machine learning. To analyze time-series changes in patents, we used the dynamic topic model (DTM), which is an extension of the latent Dirichlet allocation (LDA). Also, unlike English, the Japanese language requires morphological analysis. Patents contain many technical words that are not used in everyday life, so morphological analysis using a common dictionary is not sufficient. Therefore, we used a technique for extracting technical terms from text. After extracting technical terms, we applied them to DTM. In this study, we determined the technological progress of the lighting class F21 for 14 years and compared it with the actual revision of patent classification codes. In other words, we extracted technical terms from Japanese patents and applied DTM to determine the progress of Japanese technology. Then, we analyzed the results from the new perspective of revising patent classification codes with machine learning. As a result, it was found that those whose topics were on the rise were judged to be new technologies. △ Less

Submitted 18 December, 2020; originally announced December 2020.

Self-supervised Deep Learning for Reading Activity Classification

Authors: Md. Rabiul Islam, Shuji Sakamoto, Yoshihiro Yamada, Andrew Vargo, Motoi Iwata, Masakazu Iwamura, Koichi Kise

Abstract: Reading analysis can give important information about a user's confidence and habits and can be used to construct feedback to improve a user's reading behavior. A lack of labeled data inhibits the effective application of fully-supervised Deep Learning (DL) for automatic reading analysis. In this paper, we propose a self-supervised DL method for reading analysis and evaluate it on two classificati… ▽ More Reading analysis can give important information about a user's confidence and habits and can be used to construct feedback to improve a user's reading behavior. A lack of labeled data inhibits the effective application of fully-supervised Deep Learning (DL) for automatic reading analysis. In this paper, we propose a self-supervised DL method for reading analysis and evaluate it on two classification tasks. We first evaluate the proposed self-supervised DL method on a four-class classification task on reading detection using electrooculography (EOG) glasses datasets, followed by an evaluation of a two-class classification task of confidence estimation on answers of multiple-choice questions (MCQs) using eye-tracking datasets. Fully-supervised DL and support vector machines (SVMs) are used to compare the performance of the proposed self-supervised DL method. The results show that the proposed self-supervised DL method is superior to the fully-supervised DL and SVM for both tasks, especially when training data is scarce. This result indicates that the proposed self-supervised DL method is the superior choice for reading analysis tasks. The results of this study are important for informing the design and implementation of automatic reading analysis platforms. △ Less

Submitted 7 December, 2020; originally announced December 2020.

Comments: 28 pages

Optical vortex-induced forward mass transfer: Manifestation of helical trajectory of optical vortex

Authors: Ryosuke Nakamura, Haruki Kawaguchi, Muneaki Iwata, Akihiro Kaneko, Ryo Nagura, Satoyuki Kawano, Kohei Toyoda, Katsuhiko Miyamoto, Takashige Omatsu

Abstract: The orbital angular momentum of an optical vortex field is found to twist high viscosity donor material to form a micron-scale 'spin jet'. This unique phenomenon manifests the helical trajectory of the optical vortex. Going beyond both the conventional ink jet and laser induced forward mass transfer (LIFT) patterning technologies, it also offers the formation and ejection of a micron-scale 'spin j… ▽ More The orbital angular momentum of an optical vortex field is found to twist high viscosity donor material to form a micron-scale 'spin jet'. This unique phenomenon manifests the helical trajectory of the optical vortex. Going beyond both the conventional ink jet and laser induced forward mass transfer (LIFT) patterning technologies, it also offers the formation and ejection of a micron-scale 'spin jet' of the donor material even with an ultrahigh viscosity of 4 Pas. This optical vortex laser induced forward mass transfer (OV-LIFT) patterning technique will enable the development of next generation printed photonic/electric/spintronic circuits formed of ultrahigh viscosity donor dots containing functional nanoparticles, such as quantum dots, metallic particles and magnetic ferrite particles, with ultrahigh spatial resolution. It can also potentially explore a completely new needleless drug injection. △ Less

Submitted 9 December, 2019; originally announced December 2019.

Facial Landmark Detection for Manga Images

Authors: Marco Stricker, Olivier Augereau, Koichi Kise, Motoi Iwata

Abstract: The topic of facial landmark detection has been widely covered for pictures of human faces, but it is still a challenge for drawings. Indeed, the proportions and symmetry of standard human faces are not always used for comics or mangas. The personal style of the author, the limitation of colors, etc. makes the landmark detection on faces in drawings a difficult task. Detecting the landmarks on man… ▽ More The topic of facial landmark detection has been widely covered for pictures of human faces, but it is still a challenge for drawings. Indeed, the proportions and symmetry of standard human faces are not always used for comics or mangas. The personal style of the author, the limitation of colors, etc. makes the landmark detection on faces in drawings a difficult task. Detecting the landmarks on manga images will be useful to provide new services for easily editing the character faces, estimating the character emotions, or generating automatically some animations such as lip or eye movements. This paper contains two main contributions: 1) a new landmark annotation model for manga faces, and 2) a deep learning approach to detect these landmarks. We use the "Deep Alignment Network", a multi stage architecture where the first stage makes an initial estimation which gets refined in further stages. The first results show that the proposed method succeed to accurately find the landmarks in more than 80% of the cases. △ Less

Submitted 7 November, 2018; originally announced November 2018.

A survey of comics research in computer science

Authors: Olivier Augereau, Motoi Iwata, Koichi Kise

Abstract: Graphical novels such as comics and mangas are well known all over the world. The digital transition started to change the way people are reading comics, more and more on smartphones and tablets and less and less on paper. In the recent years, a wide variety of research about comics has been proposed and might change the way comics are created, distributed and read in future years. Early work focu… ▽ More Graphical novels such as comics and mangas are well known all over the world. The digital transition started to change the way people are reading comics, more and more on smartphones and tablets and less and less on paper. In the recent years, a wide variety of research about comics has been proposed and might change the way comics are created, distributed and read in future years. Early work focuses on low level document image analysis: indeed comic books are complex, they contains text, drawings, balloon, panels, onomatopoeia, etc. Different fields of computer science covered research about user interaction and content generation such as multimedia, artificial intelligence, human-computer interaction, etc. with different sets of values. We propose in this paper to review the previous research about comics in computer science, to state what have been done and to give some insights about the main outlooks. △ Less

Submitted 15 April, 2018; originally announced April 2018.

Heterogeneity of link weight and the evolution of cooperation

Authors: M. Iwata, E. Akiyama

Abstract: In this paper, we investigate the effect of "heterogeneity of link weight", heterogeneity of the frequency or amount of interactions among individuals, on the evolution of cooperation. Based on an analysis of the evolutionary prisoner's dilemma game on a weighted one-dimensional lattice network with "intra-individual heterogeneity", we confirm that moderate level of link-weight heterogeneity can f… ▽ More In this paper, we investigate the effect of "heterogeneity of link weight", heterogeneity of the frequency or amount of interactions among individuals, on the evolution of cooperation. Based on an analysis of the evolutionary prisoner's dilemma game on a weighted one-dimensional lattice network with "intra-individual heterogeneity", we confirm that moderate level of link-weight heterogeneity can facilitate cooperation. Furthermore, we identify two key mechanisms by which link-weight heterogeneity promotes the evolution of cooperation: mechanisms for spread and maintenance of cooperation. We also derive the corresponding conditions under which the mechanisms can work through evolutionary dynamics. △ Less

Submitted 31 October, 2015; v1 submitted 24 July, 2015; originally announced July 2015.

Comments: 35 pages, 37 figures, Submitted to Physica A

Theoretical analysis for critical fluctuations of relaxation trajectory near a saddle-node bifurcation

Authors: Mami Iwata, Shin-ichi Sasa

Abstract: A Langevin equation whose deterministic part undergoes a saddle-node bifurcation is investigated theoretically. It is found that statistical properties of relaxation trajectories in this system exhibit divergent behaviors near a saddle-node bifurcation point in the weak-noise limit, while the final value of the deterministic solution changes discontinuously at the point. A systematic formulation f… ▽ More A Langevin equation whose deterministic part undergoes a saddle-node bifurcation is investigated theoretically. It is found that statistical properties of relaxation trajectories in this system exhibit divergent behaviors near a saddle-node bifurcation point in the weak-noise limit, while the final value of the deterministic solution changes discontinuously at the point. A systematic formulation for analyzing a path probability measure is constructed on the basis of a singular perturbation method. In this formulation, the critical nature turns out to originate from the neutrality of exiting time from a saddle-point. The theoretical calculation explains results of numerical simulations. △ Less

Submitted 30 June, 2010; v1 submitted 22 February, 2010; originally announced February 2010.

Comments: 18pages, 17figures.The version 2, in which minor errors have been fixed, will be published in Phys. Rev. E

Journal ref: Phys. Rev. E. 82, 011127 (2010)

Singular perturbation near mode-coupling transition

Authors: Mami Iwata, Shin-ichi Sasa

Abstract: We study the simplest mode-coupling equation which describes the time correlation function of the spherical p-spin glass model. We formulate a systematic perturbation theory near the mode-coupling transition point by introducing multiple time scales. In this formulation, the invariance with respect to the dilatation of time in a late stage yields an arbitrary constant in a leading order expressi… ▽ More We study the simplest mode-coupling equation which describes the time correlation function of the spherical p-spin glass model. We formulate a systematic perturbation theory near the mode-coupling transition point by introducing multiple time scales. In this formulation, the invariance with respect to the dilatation of time in a late stage yields an arbitrary constant in a leading order expression of the solution. The value of this constant is determined by a solvability condition associated with a linear singular equation for perturbative corrections in the late stage. The solution thus constructed provides exactly the alpha-relaxation time. △ Less

Submitted 28 May, 2009; v1 submitted 5 February, 2009; originally announced February 2009.

Comments: 15 pages, 4 figures

Journal ref: J. Phys. A: Math. Theor. 42 (2009) 245001

Dynamics of k-core percolation in a random graph

Authors: Mami Iwata, Shin-ichi Sasa

Abstract: We study the edge deletion process of random graphs near a k-core percolation point. We find that the time-dependent number of edges in the process exhibits critically divergent fluctuations. We first show theoretically that the k-core percolation point is exactly given as the saddle-node bifurcation point in a dynamical system. We then determine all the exponents for the divergence based on a u… ▽ More We study the edge deletion process of random graphs near a k-core percolation point. We find that the time-dependent number of edges in the process exhibits critically divergent fluctuations. We first show theoretically that the k-core percolation point is exactly given as the saddle-node bifurcation point in a dynamical system. We then determine all the exponents for the divergence based on a universal description of fluctuations near the saddle-node bifurcation. △ Less

Submitted 21 January, 2009; v1 submitted 5 August, 2008; originally announced August 2008.

Comments: 16 pages, 4 figures

Journal ref: J. Phys. A: Math. Theor. 42 (2009) 075005

Scale-free patterns at a saddle-node bifurcation in a stochastic system

Authors: Mami Iwata, Shin-ichi Sasa

Abstract: We demonstrate that scale-free patterns are observed in a spatially extended stochastic system whose deterministic part undergoes a saddle-node bifurcation. Remarkably, the scale-free patterns appear only at a particular time in relaxation processes from a spatially homogeneous initial condition. We characterize the scale-free nature in terms of the spatial configuration of the exiting time from… ▽ More We demonstrate that scale-free patterns are observed in a spatially extended stochastic system whose deterministic part undergoes a saddle-node bifurcation. Remarkably, the scale-free patterns appear only at a particular time in relaxation processes from a spatially homogeneous initial condition. We characterize the scale-free nature in terms of the spatial configuration of the exiting time from a marginal saddle where the pair annihilation of a saddle and a node occurs at the bifurcation point. Critical exponents associated with the scale-free patterns are determined by numerical experiments. △ Less

Submitted 14 November, 2008; v1 submitted 6 June, 2008; originally announced June 2008.

Comments: 4 pages, 6 figures

Journal ref: Phys. Rev. E 78, 055202(R) (2008)

A theory for critically divergent fluctuations of dynamical events at non-ergodic transitions

Authors: Mami Iwata, Shin-ichi Sasa

Abstract: We theoretically study divergent fluctuations of dynamical events at non-ergodic transitions. We first focus on the finding that a non-ergodic transition can be described as a saddle connection bifurcation of an order parameter for a time correlation function. Then, following the basic idea of Ginzburg-Landau theory for critical phenomena, we construct a phenomenological framework with which we… ▽ More We theoretically study divergent fluctuations of dynamical events at non-ergodic transitions. We first focus on the finding that a non-ergodic transition can be described as a saddle connection bifurcation of an order parameter for a time correlation function. Then, following the basic idea of Ginzburg-Landau theory for critical phenomena, we construct a phenomenological framework with which we can determine the critical statistical properties at saddle connection bifurcation points. Employing this framework, we analyze a model by considering the fluctuations of an instanton in space-time configurations of the order parameter. We then obtain the exponents characterizing the divergences of the length scale, the time scale and the amplitude of the fluctuations of the order parameter at the saddle connection bifurcation. The results are to be compared with those of previous studies of the four-point dynamic susceptibility at non-ergodic transitions in glassy systems. △ Less

Submitted 24 January, 2007; v1 submitted 11 September, 2006; originally announced September 2006.

Comments: 5 pages. Minor revision. This version will appear in Europhys. Lett

Journal ref: EPL 77 (2007) 50008

Non-ergodic transitions in many-body Langevin systems: a method of dynamical system reduction

Authors: Mami Iwata, Shin-ichi Sasa

Abstract: We study a non-ergodic transition in a many-body Langevin system. We first derive an equation for the two-point time correlation function of density fluctuations, ignoring the contributions of the third- and fourth-order cumulants. For this equation, with the average density fixed, we find that there is a critical temperature at which the qualitative nature of the trajectories around the trivial… ▽ More We study a non-ergodic transition in a many-body Langevin system. We first derive an equation for the two-point time correlation function of density fluctuations, ignoring the contributions of the third- and fourth-order cumulants. For this equation, with the average density fixed, we find that there is a critical temperature at which the qualitative nature of the trajectories around the trivial solution changes. Using a method of dynamical system reduction around the critical temperature, we simplify the equation for the time correlation function into a two-dimensional ordinary differential equation. Analyzing this differential equation, we demonstrate that a non-ergodic transition occurs at some temperature slightly higher than the critical temperature. △ Less

Submitted 31 August, 2006; v1 submitted 2 May, 2006; originally announced May 2006.

Comments: 8 pages, 1 figure; ver.3: Calculation errors have been fixed

Journal ref: J. Stat. Mech. (2006) L10003

Atomic Carbon and CO Isotope Emission in the Vicinity of DR15

Authors: T. Oka, S. Yamamoto, M. Iwata, H. Maezawa, M. Ikeda, T. Ito, K. Kamegai, T. Sakai, Y. Sekimoto, K. Tatematsu, Y. Arikawa, Y. Aso, T. Noguchi, S-C. Shi, K. Miyazawa, S. Saito, H. Ozeki, H. Fujiwara, M. Ohishi, J. Inatani

Abstract: We present observations of the 3P1-3P0 fine structure transition of atomic carbon [CI], the J=3-2 transition of CO, as well as of the J=1-0 transitions of 13CO and C18O toward DR15, an HII region associated with two mid-infrared dark clouds (IRDCs). The 13CO and C18O J=1-0 emissions closely follow the dark patches seen in optical wavelength, showing two self-gravitating molecular cores with mass… ▽ More We present observations of the 3P1-3P0 fine structure transition of atomic carbon [CI], the J=3-2 transition of CO, as well as of the J=1-0 transitions of 13CO and C18O toward DR15, an HII region associated with two mid-infrared dark clouds (IRDCs). The 13CO and C18O J=1-0 emissions closely follow the dark patches seen in optical wavelength, showing two self-gravitating molecular cores with masses of 2000 Msun and 900 Msun, respectively, at the positions of the catalogued IRDCs. Our data show a rough spatial correlation between [CI] and 13CO J=1-0. Bright [CI] emission occurs in relatively cold gas behind the molecular cores, neither in highly excited gas traced by CO J=3-2 emission nor in HII region/molecular cloud interface. These results are inconsistent with those predicted by standard photodissociation region (PDR) models, suggesting an origin for interstellar atomic carbon unrelated to photodissociation processes. △ Less

Submitted 30 May, 2001; originally announced May 2001.

Comments: 11 pages Latex, 6 figures, Accepted for publication in The Astrophysical Journal

Journal ref: Astrophys.J. 623 (2005) 889-896