Suitable collective decision making is crucial both in real life and in the field of computer science. Previous studies have implemented two systems using photons with superposition of orbital angular momenta (OAMs) for collective decision making: one that allows only symmetric treatments of agents and another that enables asymmetric treatments. However, these systems cannot assemble all probability matrices, i.e., only limited stochastic decision making was possible. Therefore, this study introduces a way called the enlarged matrix method to diversify stochastic decision making. The enlarged matrix method successfully extends the range of possible probability matrices, indicating the benefit of utilizing photons with superposed OAMs for various stochastic decision making. Also, this study applies decision-making systems utilizing OAM to correlated equilibria, a game theory concept that preserves benefits for all players by repressing selfish decisions. Possible equilibria for each system are described numerically and theoretically.

This paper describes a machine learning of image reconstruction for computed tomography. This method is a projection-based nonlinear model order reduction algorithm for a nonlinear dynamical system. The projection matrix is obtained from a weighted variance-covariance matrix, the so-called snapshot matrix. The learning is done so that the projection matrix can compactly represent many projection data acquired from X-ray sensors. After learning the nonlinear dynamics, high-quality reconstruction images are very efficiently obtained by solving a least-squares equation. The developed image reconstruction system is effective for low X-ray computed tomography apparatus, which benefits patients.

In emergencies such as disasters, the number of voice calls (VoIP sessions) increases rapidly for a variety of purposes. Thus, a control server near a disaster area may not be able to connect to VoIP sessions due to congestion. To solve this problem, a Call Admission Control (CAC) is needed to determine whether a VoIP session requesting a connection can be accepted or rejected. A CAC has the purpose of guaranteeing the connection quality and communication quality of VoIP sessions. One conventional method classifies VoIP sessions into three classes (emergency VoIP sessions, VoIP sessions from the disaster area, and VoIP sessions from outside the disaster area) by focusing on the outgoing location and offers a CAC with a priority level for each. However, a conventional CAC method cannot be applied to VoIP networks because reception control is designed for Public Switched Telephone Networks (PSTN). When conventional methods are applied to VoIP networks, the connection quality is guaranteed, however the communication quality cannot be guaranteed because the packet dropping probability is not considered. In this paper, we propose a three-dimensional CAC that controls three classes of VoIP sessions and guarantees both communication and connection quality in VoIP networks during emergencies. A conventional CAC method and our proposed CAC method are evaluated in terms of the call blocking probability, which guarantees the connection quality, and packet dropping probability, which guarantees the communication quality, to show the effectiveness of the proposed method.

Image inpainting aims to recover or reconstruct specific regions in given images, and it is also used to remove unwanted objects in the images. Existing image inpainting methods can be classified into patch-based methods and machine learning-based methods. The former suffers from a lack of global consistency and diversity in the restored region, while the latter is challenged by its dependence on the dataset domain used for training. On the other hand, single image generative adversarial networks (SinGAN) has been proposed to generate a variety of images from a given single image, which is a dataset-independent method because it does not require prior training. However, SinGAN cannot be applied to image inpainting because it requires defect-free images for training. In this study, we propose a globally consistent, dataset-independent image inpainting method with diverse outputs. Our proposed method generalizes the architecture of SinGAN to be used for image inpainting by introducing partial convolution and region normalization to train the generative neural network from a single image containing unwanted or missing regions. To confirm the effectiveness of the proposed method, we quantitatively and qualitatively compared its image inpainting performance with existing patch-based methods. Experimental results show that the proposed method can perform image inpainting with global consistency and diversity while the restoration performance is comparable to that of existing methods.

High-speed ship operations have a significant risk of the broaching phenomenon when navigating in following/quartering seas. The occurrence of this phenomenon can result in a violent yaw motion, regardless of the steering effort. Centrifugal forces due to the yaw motion could cause large roll motion and even capsizing in some cases. A necessary condition for the occurrence of broaching is the surf-riding phenomenon. Therefore, the International Maritime Organization (IMO) has set up criteria to include theoretical formulas for estimating the occurrence of surf-riding phenomena. The theoretical equation used in the IMO Second-Generation Intact Stability criteria (SGISC) to estimate the surf-riding threshold is based on the authors' method, and in its derivation process, Melnikov's method. is utilized. This paper deals with nonlinear equations describing the surge motions of a ship. However, such equations cannot be directly solved; therefore, prior researchers and authors have proposed various approximate methods to predict the surf-riding threshold, including the methods based on Meknikov's method. In this paper, first, the authors describe the mechanism of the surfing phenomenon. Then, the authors explain in detail the derivation of the estimation methods for the surf-riding phenomenon that have been proposed by the prior researchers and the authors. Finally, the authors show the relationship between the theoretical prediction method of the surf-riding threshold based on Melnikov's method and the IMO SGISC.

Discrete breathers are spatially localized time-periodic solutions in nonlinear lattices. We study the scattering of small amplitude phonons by a discrete breather in two types of one-dimensional nonlinear lattices, which are called the pairwise interaction symmetric lattice (PISL) and the Umklapp-free lattice (UFL). Each lattice has its own particular symmetry in the potential function and is known to exhibit quite anomalous heat transport: the PISL exhibits an almost ballistic transport;the UFL the ballistic transport implying no thermal resistance. We numerically calculate the transmission and reflection rates of a phonon wave packet as a function of the wavenumber. It is shown that almost perfect transmission occurs in the PISL while almost perfect reflection in the UFL.

Nurmuhammad et al. developed the Sinc-Nyström methods for initial value problems in which the solutions exhibit exponential decay end behavior. In these methods, the Single-Exponential (SE) transformation or the Double-Exponential (DE) transformation is combined with the Sinc approximation. Hara and Okayama improved on these transformations to attain a better convergence rate, which was later supported by theoretical error analyses. However, these methods have a computational drawback owing to the inclusion of a special function in the basis functions. To address this issue, Okayama and Hara proposed Sinc-collocation methods, which do not include any special function in the basis functions. This study conducts error analyses of these methods.