首页|Investigators from National Nanotechnology Center Zero in on Machine Learning (E xpanding the Applicability Domain of Machine Learning Model for Advancements In Electrochemical Material Discovery)
Investigators from National Nanotechnology Center Zero in on Machine Learning (E xpanding the Applicability Domain of Machine Learning Model for Advancements In Electrochemical Material Discovery)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Investigators publish new report on Ma chine Learning.According to news reporting originating in Pathum Thani,Thailan d,by NewsRx journalists,research stated,"Machine learning has gained consider able attention in the material science domain and helped discover advanced mater ials for electrochemical applications.Numerous studies have demonstrated its po tential to reduce the resources required for material screening." Financial supporters for this research include VISTEC-NSTDA collaborative resear ch and education scholarship,VISTEC-NSTDA collaborative research and education scholarship,National Research Council of Thailand (NRCT),Program Management Un it for Human Resources & Institutional Development,Research and I nnovation.The news reporters obtained a quote from the research from National Nanotechnolo gy Center,"However,a significant proportion of these studies have adopted a su pervised learning approach,which entails the laborious task of constructing ran dom training databases and does not always ensure the model's reliability while screening unseen materials.Herein,we evaluate the limitations of supervised ma chine learning from the perspective of the applicability domain.The applicabili ty domain of a model is the region in chemical space where the structure-propert y relationship is covered by the training set so that the model can give reliabl e predictions.We review methods that have been developed to overcome such limit ations,such as the active learning framework and self-supervised learning.The effort required for material discovery has decreased thanks to machine learning.However,the quality of the dataset,which should be vast and diverse,determin es the model's effectiveness and reliability.Therefore,the trustworthiness of prediction should be evaluated based on the concept of applicability domain."
Pathum ThaniThailandAsiaChemicalsCyborgsElectrochemicalsEmerging TechnologiesMachine LearningSupervised LearningNational Nanotechnology Center
NETL
NSTL
