Close association between intestinal microbiota and the immune system suggests possibility of developing a novel treatment method for allergy by modulating intestinal microbiota. Although time-series data with sufficient length are required for microbiota modulation, no such datasets are found which match the requirement. In this study, we apply and examine a new method to reconstruct a pseudo-temporal path of allergy development from non-time-series samples reflecting various individual states in the process of allergy development. We aim to refine this method to provide a data-driven intervention strategy for prevention of allergy.
The purpose of this paper is to introduce various nonnegative systems and preset the stability analysis method of each system. Biological phenomena are modeled using nonnegative conditions such as the number of animals, the concentration of chemicals in the body. Nonnegative systems represent not only biological and biological phenomena but also various phenomena. This paper introduces various nonnegative systems such as nonnegative linear system, Lotka-Volterra equation, positive secondary system, and so on. In addition, this paper clarifies the range of phenomena that each nonnegative system can express. Finally, this paper introduces the stability analysis method peculiar to the nonnegative system.
In these decades, high-frequency financial data, which are big data containing detailed information on each transaction, have become available, and there has been growing needs for methods to analyze such data. Recently, point processes have been recognized as a useful tool to analyze high-frequency data. Point processes are probability models of timings of occurrences of events, and Hawkes processes, a type of point processes, can reproduce actually observed occurrence patterns of transactions or orders. In this article, we introduce a basic concept of point processes, and review our recent study on high-frequency financial data analysis using Hawkes processes.
We develop the theoretical foundation on the proximal method of multipliers for a class of nonsmooth convex optimization. The method generates a sequence of subproblems for the objective functions of the sum of the augmented Lagrangian due to Fortin and the proximal term. We show that the sequence of approximations to the subproblems converges to a saddle point of the standard Lagrangian even if the original optimization problem may possess multiple solutions. We employ a nonsmooth Newton method for computing an approximation to the subproblem. We exploit the theory of the nonsmooth analysis to provide a rigorous proof for the global convergence of the nonsmooth Newton algorithm.
This review article introduces timeseries analysis based on an aggregating algorithm Expert Advice. In the Expert Advice, an expert possesses its weight that is determined by its accumulated loss of errors. Then, the prediction of the Expert Advice is obtained by the weighted average over the predictions of these experts and their weights. This review article gives theoretical introduction and examples of application of the Expert Advice.
In recent years, an integrated circuit （IC） chip development for an artificial neural network （ANN） is accelerated. Power consumption for a multiply-accumulate operation of an analog ANN chip is lower than that of a digital ANN. However, the analog ANN strongly influenced by the process variations. In order to confirm whether that an analog ANN chip without compensation can on-chip learning in real time, I have been developing an ultra-low power analog CMOS spiking neural network LSI chip. In this report, I show circuit simulation results of its synapse circuit.
Noise-induced synchronization is a phenomenon that the nonlinear oscillators synchronize by adding common noise to each of them. A new synchronization scheme based on noise-induced synchronization phenomenon has been proposed, and uncoupled wireless devices are synchronized without any wireless communication by the scheme. In the proposed scheme, natural fluctuations are applied to the nonlinear oscillators as common noise of noise-induced synchronization phenomenon. This paper explains detail of this natural synchronization scheme. This paper gives a brief exposition about confirmation of synchronization phenomenon by theory and numerical simulations and verifying synchronization performance by an implementation to wireless sensor network and so on.
A deep neural network (DNN) has been in the center of attention in the field of machine learning, and the chaotic neuron and chaotic neural network (ChNN) models have been in the spotlight in computational neuroscience and nonlinear science. However, there are no studies on deep ChNN. In order to fill this gap, we propose a ReLU (rectifier linear unit) chaotic neuron model, which is necessary for the application of the chaotic neuron model to DNN with ReLU activation. We also show that even a single ReLU chaotic neuron can generate dynamically changing outputs in spite of the simplicity of ReLU.
SMMiL-E is the first overseas laboratory of the Institute of Industrial Science, the University of Tokyo.
In SMMiL-E, interdisciplinary and international collaborations are preformed to implement microsystems developed in LIMMS at the IIS into cancer therapies in Lille, France. To proceed with these collaborations smoothly and fruitfully, a new function called “Scientific Direction” has been implemented. Specifically, two medical doctors, one French and the other Japanese, have been selected and assigned as Scientific Directors to supervise and guide the research within SMMiL-E. Scientific Direction is now expected to expand to uphold the qualities of the collaborative projects in SMMiL-E, as the organization continues to grow in scale. Efficient Scientific Direction is a key to the success of the SMMiL-E endeavour.
The development of high-throughput and accurate analytical methods is highly important from the viewpoint of environmental monitoring or medicinal applications. In this regard, much attention has been paid to chemosensor arrays inspired by a mammalian olfactory system, because they can offer the simultaneous detection of various analytes. However, high effort is required for the synthesis of chemosensors, which might prevent generalization and mass production in industry. Toward that end, we propose an easy-to-prepare colorimetric chemosensor array only using commercially available materials. As an actual approach, the simultaneous detection of metal ions is herein reported.
Transition metal compounds play an important role as the catalyst for the various organic transformation reactions in both laboratory experiments and industrial applications. It is well known that the ligand exchange is the primary step in many cases of these catalysis. In this paper, the ligand exchange reaction of the CNtBu ligands on the planar tetranuclear palladium cluster, Pd4[Si(iPr2)]3(CNtBu)4 (1), to the iPr2IMMe ligands have been elucidated. In the course of this reaction, formation of a new Pd4 cluster Pd4[Si(iPr2)]3(CNtBu)2 (iPr2IMMe)3 (3) as intermediary species was confirmed, suggesting that ligand exchange reaction proceeded through the associative mechanism.
We observe the permeability of several commercially-available microneedle patches for cosmetic purposes to human skin (back of hand), and investigate methods of evaluating permeability to the stratum corneum, and its’ degree of pain. We predicted that the permeability of dissoluble microneedles differs according to the age and sex of humans, then we evaluated the permeability based on 18 person’s clinical tests by using imagediagnosis of“ optical coherence tomography (OCT)”. The basic data obtained in this research can be used for developing an experimental plan for new microneedle fabrication to be implemented in further medical applications.
Blood extracting Microneedle has been studied for minimally invasive and rapid blood tests. However, in practical use of solid micro needles, the risk is concerned that microneedle can be easily broken due to the load of puncture and that of remaining broken needle in the subcutaneous. To solve the risk, this research has developed a blood extracting microneedle using biodegradable polymer. In this research, instead of the conventional hollow needle, we proposed the porous polymer needles to absorb blood. For the development of porous needles, experiments were conducted to determine suitable porosity. As a result, 60% or more of porosity is suitable for blood extracting and its minimum extracting speed is 1.75 μl/min.
Continuing from our previous paper, in this section, we investigated the following three items in the Web of Science of academic literature search service: (i) the number of documents including ‘structure’ in their titles, (ii) in what research field does the literature includes ‘structure’ in the title, and (iii) what kind of research fields overlap for literature that include‘ structure’ in the title. We obtained the knowledge that the word of ‘structure’ was mainly used in the field of ‘Physical Sciences’. In recent years it was also used in the field of ‘Technology’.
Silicon nanotweezers (SNT) are a mechanical characterization tool for biological samples such as DNA bundles or single cells. With this device, it is possible to perform mechanical characterization followed by genetic analysis of a single cell. However, it is difficult to release cells from the SNT probes after mechanical characterization because cells tend to physically adhere to the silicon or glass surface of the device.
To solve this problem, we explored surface modification techniques for the SNT probes using several types of chemical coatings as well as modifying the scallop size on the silicon sidewall during fabrication. We measured the hydrophobicity of modified silicon surfaces, then treated SNT probes and evaluated their performance in cell trapping and releasing.