The University of Osaka · 공학
와시오 타카시 교수의 연구실은 데이터 기반 화학 합성 및 분석을 핵심으로 하며, 특히 기계학습과 최적화 기법을 활용한 반응 조건 스크리닝, 전기화학적 유량 반응기에서의 다중 매개변수 최적화, 단일 분자 측정의 정밀 분석 기술 개발에 주력하고 있습니다. 그래프 기반 데이터 마이닝, 스케일 타입 기반 방정식 발견 시스템(SDS), 그리고 고자기 선택성 반응의 유기촉매적 합성 등 다양한 분야에서 이론적 기반과 응용 기술을 융합한 연구를 수행하고 있습니다. 특히, 반응 조건의 자동 최적화와 단일 분자 수준의 정밀 측정 기술을 통해 화학 합성의 효율성과 정밀도를 극대화하는 데 초점을 맞추고 있습니다.
표시된 성과는 수집된 데이터 기준으로 산출되며, 일부 차이가 있을 수 있습니다.
The need for mining structured data has increased in the past few years. One of the best studied data structures in computer science and discrete mathematics are graphs. It can therefore be no surprise that graph based data mining has become quite popular in the last few years.This article introduces the theoretical basis of graph based data mining and surveys the state of the art of graph-based data mining. Brief descriptions of some representative approaches are provided as well.
A highly atom-economical enantioselective organocatalyzed Rauhut-Currier and [3+2] annulation sequence has been established by using a flow system. Suitable flow conditions were explored through reaction screening of multiple parameters using machine learning. Eventually, functionalized chiral spirooxindole analogues were obtained in high yield with good ee as a single diastereomer within one minute.
When conducting a single-molecule measurement, data are analyzed using the histogram of a measured physical quantity in which a single dataset contains information about a specific single molecule. Oftentimes, the histogram consisting of only specific single-molecule information excludes the input from other information sources. In other words, despite measuring the single molecule during analysis, we miss many of the properties of that single molecule. Herein, we have successfully identified a
Multiparameter screening of reductive carboxylation in an electrochemical flow microreactor was performed using a Bayesian optimization (BO) strategy. The developed algorithm features a constraint on passed charge for the electrochemical reaction, which led to suitable conditions being instantaneously found for the desired reaction. Analysis of the BO-suggested conditions underscored the physicochemical validity.
SDS is a discovery system from numeric measurement data. It outperforms the existing systems in every aspect of search e ciency, noise tolerancy, credibility of the resulting equations and complexity of the target system that it can handle. The power of SDS comes from the use of the scale-types of the measurement data and mathematical property of identity by which to constrain the admissible solutions. Its algorithm is described with a complex working example and the performance comparison with
A highly efficient synthesis of α-ketiminophosphonates has been established for the electrochemical oxidation of α-amino phosphonates with the utilization of machine-learning-assisted simultaneous multiparameter screening.
The variability of bioparticles remains a key barrier to realizing the competent potential of nanoscale detection into a digital diagnosis of an extraneous object that causes an infectious disease. Here, we report label-free virus identification based on machine-learning classification. Single virus particles were detected using nanopores, and resistive-pulse waveforms were analyzed multilaterally using artificial intelligence. In the discrimination, over 99% accuracy for five different virus sp
Most conventional law equation discovery systems suchasBACON require experimental environments to acquire their necessary data. The mathematical techniques such as linear system identification and neural network fitting presume the classes of equations to model given observed data sets. The study reported in this paper proposes a novel method to discover an admissible model equation from a given set of observed data, while the equation is ensured to reflect first principles governing the objecti
