Kyutacar : Kyushu Institute of Technology Academic Repository
Not a member yet
9746 research outputs found
Sort by
Complementary Duals—Both Fixed Points (III)—
Get PDFWe consider a quadratic minimization problem with both fixed endpoints and its associated maximization problem from a viewpoint of complementarity. We show that a complementary identity with an elementary inequality generates a pair of dn _ 1T-variable minimization problem (primal) and dn t 1T-variable maximization one (dual). The identity produces an equality condition, which is a linear system of 2n-equation on 2n-variable. The condition splits into a pair of linear systems of dn _ 1T-equation on dn _ 1T-variable and of dn t 1T-equation on dn t 1T-variable. The former solves the primal, while the latter does the dual. Both the optimal solutions are characterized by the Fibonacci sequence. The solutions are also given through dynamic programming.departmental bulletin pape
When LoRaWAN Meets CSMA: Trends, Challenges, and Opportunities
Get PDFLong-Range Wide-Area Network (LoRaWAN) has been increasingly deployed to serve as the wireless networking platforms for various Internet-of-Things applications. However, an inherent drawback of LoRaWAN is that the signal transmissions of LoRaWAN end devices frequently collide with each other, which leads to data decoding failure. This is because LoRaWAN enables wireless channel access mainly based on the naive ALOHA protocol where each end device is allowed to transmit regardless of others' ongoing transmissions. In this context, the past few years have witnessed the prosperity of enhanced channel access protocol design for LoRaWAN. Particularly, carrier-sense multiple access (CSMA) has gained more attention recently since it enables channel access in a distributed manner without additional signaling overhead between LoRaWAN end devices and gateways. More importantly, implementing CSMA in LoRaWAN is a challenging and interesting problem since signals in LoRaWAN can traverse below the noise floor, making traditional carrier-sensing techniques fail to detect ongoing transmissions. By considering the necessity and importance of this topic, this article aims to provide a comprehensive and timely review of CSMA-based LoRaWAN for the first time. We start by analyzing the trends of existing LoRaWAN-oriented CSMA studies. As the key possible technique to enable CSMA in practical LoRaWAN, Channel Activity Detection (CAD) and its associated challenges are then deeply investigated with a real-world measurement study. Based on the limitations of CAD, we finally provide several research insights for CAD-based CSMA in LoRaWAN.journal articl
Practice of Career Education Using ICT in the Period for Integrated Studies at Junior High School: A Case Study of Remote Joint Classes within a School
Get PDFThis study reports a case study on career education practice using ICT in integrated studies at a junior high school. Learning activities in the joint classes included collaborative learning and research using tablets. The analysis examined the process of transformation of students’ learning. The results suggest that student learning was deepened by career education practice.departmental bulletin pape
Peculiar spin glass phase emerging in FeCo/FePt driven via nanoconfined crystallographic distortions
Get PDFWe explore the existence of spin glass phase in FeCo/FePt bilayers arising due to disordered ferromagnet. The non-ergodic and highly degenerate landscape of the spin glass phase at low temperature explains the origin of complex magnetic texture in the FeCo/FePt system. Upon cooling the bilayered system, the magnetic texture undergoes spin freezing below 120 K as evident from the bifurcations in zero field cooling and field cooling magnetizations at low magnetic field as a manifestation of broken ergodicity. The uncompensated magnetic moments originating in the spin glass state result in slow time dynamics of thermoremanent magnetization. Consequently, the bilayers demonstrate an intriguing magnetic memory effect in which the magnetic state of the system could be retrieved upon isothermal ageing below 120 K after reversing the temperature cycle. Thermal treatment deteriorates the spin glass behaviour and shows a transition to strong ferromagnetic character in FeCo/FePt bilayers.journal articl
Analytical Development Method for Object Detection Dataset
Get PDF本研究では，機械学習に用いたときに，人が作成するデータセットと同等の精度を得られるデータセットを自動生成する手法の実現を目指す．特に画像を解析しながらデータセットを構築する手法に注目して取り組む．具体的には，生成に用いる画像を解析し，生成方法を設計，データセット生成後に生成したデータセットを解析する，というものである．本稿ではこの一連の処理に対してトイデータや軽量なアルゴリズムを採用し，まずはツールチェーンの構築に取り組む．評価実験では，データセットの解析値や学習した結果を使って定量的・定性的にツールの機能や提案手法の学習への効果を確認した．We propose an automatic dataset generation method which creates image with analyzing image, in order to yield a dataset that give the same performance as a hand-made dataset to an machine learning algorithm. The proposed method analyze component of dataset generation, design how to generate, and analyze a generated dataset. This paper adopt toy data and a basic algorithm due to construct the whole proposed processes easily, for the first step of this research. Experimental results reports functionality and effect of the proposed method, both quantitatively and qualitatively, using the analysis values of the dataset and the results of the training.journal articl
Some Implications of Positivity with Interior Equilibria via Asymmetrically Scaled Sectorial Supply Rates
Get PDFThis paper presents some results to develop dissipative systems theory in terms of asymmetrically scaled sectorial supply rates. The supply rates, which describe strict passivity on asymmetric spaces, were proposed recently by the author to simultaneously verify stability and positivity of dynamical systems whose equilibria are interior points of positive state spaces. This paper demonstrates that systems defined by an asymmetrically scaled sectorial supply rate have finite asymmetric gain if their positivity is guaranteed by the supply rate. For cyclic networks whose component systems are defined by asymmetrically scaled sectorial supply rates, this paper shows that the networks are globally asymptotically stable on the positive space whenever positivity is guaranteed by the individual supply rates. The developments deal with general nonlinear systems. It is illustrated that positive systems design benefits a lot from asymmetrically scaled sectorial supply rates and their implications even for affine compartmental systems for which symmetric methods are ineffective.journal articl
Experimental Comparison of Human-based Evolutionary Computation and ChatGPT
Get PDFIn recent years, generative AI, which is a mechanism for automatically generating content that meets human needs, has been attracting attention. There are a wide variety of types of generated content, and ChatGPT is a typical example of a generated AI that responds to requests from humans using natural language. Generative AI acquires generation methods based on automatic learning of past data. On the other hand, people have been gathering their knowledge, experience, and creativity to solve various problems on the Web. Q&A sites are a typical example. Furthermore, in the field of evolutionary computation, there is human-based evolutionary computation, in which a group of humans solves problems by performing evolutionary computation. Now that generative AI has appeared, it is necessary to identify problems for which problem-solving methods based solely on the power of human groups are effective. Therefore, in this study, we compare the problem-solving performance of ChatGPT, a generative AI, and human-based evolutionary computation based only on people’s power, for two types of problems. The two types of problems are problems for which the Web is full of information that can help solve problems, and problems for which there is little such problem on the Web. The experimental results suggest that ChatGPT is more suitable for the former type of problem and the human-based evolutionary computation is more suitable for the latter type of problem.journal articl
Self-referential holography and its applications to data storage and deep neural network hardware
Get PDFHolography is a technology that enables the recording of light waves. To record a hologram, in addition to the object light that the beam to be recorded, a reference beam is required. However, the use of a reference beam can lead to challenges such as increased system size and complexity in holographic applications.conference pape
Numerical simulation of deep reinforcement learning using self-referential holographic deep neural network
Get PDFDeep reinforcement learning (DRL) is an artificial intelligence that is capable of autonomous decision-making in complex situations through a complementary combination of deep neural networks (DNNs) and reinforcement learning (RL). The autonomy is based on the cooperation between RL, which explores adaptations through trial and error in the environment; database (DB), which stores experience; and DNN, which learns adaptive behavior from training on experience data. The cooperation has been successfully applied to tasks that must operate directly in the face of a complex world, such as robotics, Go AI, and data center cooling. Regarding DNNs, the advancement of applications and algorithms requires the improvement of energy efficiency. There are two approaches to address this issue. The first is based on the current mainstream of electronic digital computing, such as application-specific integrated circuits (ASICs) and field-programmable gate arrays (FPGAs). The second explores new computing paradigms not dependent on von Neumann architectures, such as the search for extreme performance using optical computing. Although ASIC- and FPGA-based approaches have been studied for DNNs used in DRLs named q-networks [1,2], to the best of the authors' knowledge, there is no research on non-von Neumann computing. Here, we focus on self-referential holography (SRH) [3], an optoelectronic technology that implements two functions, as large-scale DB and DNNs exploiting the spatial parallelism of light, in a single system. In this study, we numerically investigate the feasibility of q-networks based on self-referential holographic DNN (SR-HDNN) [4].conference pape