Machine learning (ML) has been widely used todesign and develop new materials owing to its low computational cost and powerful predictive capabilities. In recentyears, the shortcomings of ML in materials science have ...Machine learning (ML) has been widely used todesign and develop new materials owing to its low computational cost and powerful predictive capabilities. In recentyears, the shortcomings of ML in materials science have gradually emerged, with a primary concern being the scarcity ofdata. It is challenging to build reliable and accurate ML modelsusing limited data. Moreover, the small sample size problemwill remain long-standing in materials science because of theslow accumulation of material data. Therefore, it is importantto review and categorize strategies for small-sample learningfor the development of ML in materials science. This reviewsystematically sorts the research progress of small-samplelearning strategies in materials science, including ensemblelearning, unsupervised learning, active learning, and transferlearning. The directions for future research are proposed, including few-shot learning, and virtual sample generation.More importantly, we emphasize the significance of embedding material domain knowledge into ML and elaborate on thebasic idea for implementing this strategy.展开更多
publishes only papers of the highest quality. All papers submitted are sent to referees familiar with the respective subjects for advice as to their eligibility for publication and needed improvement. The Editors make...publishes only papers of the highest quality. All papers submitted are sent to referees familiar with the respective subjects for advice as to their eligibility for publication and needed improvement. The Editors make the final decision on publication in the light of such advice. Non-native English-speaking authors are advised to seek necessary linguistic assistance before submission of their manuscripts,although the Editors are prepared to help upgrade works of exceptional excellence. In addition to full length papers of up to 5000 words, the journal welcomes reviews, correspondence and shorter communications of less than 2500 words dealing with novel research techniques, analysis of current research, new concepts, etc.展开更多
基金supported by the National Natural Science Foundation of China (52371007 and 52301042)the National Key R&D Program of China (2020YFB0704503)+1 种基金the Guangdong Basic and Applied Basic Research Foundation (2021B1515120071)the Key-Area Research and Development Program of Guangdong Province (2023B0909050001)。
文摘Machine learning (ML) has been widely used todesign and develop new materials owing to its low computational cost and powerful predictive capabilities. In recentyears, the shortcomings of ML in materials science have gradually emerged, with a primary concern being the scarcity ofdata. It is challenging to build reliable and accurate ML modelsusing limited data. Moreover, the small sample size problemwill remain long-standing in materials science because of theslow accumulation of material data. Therefore, it is importantto review and categorize strategies for small-sample learningfor the development of ML in materials science. This reviewsystematically sorts the research progress of small-samplelearning strategies in materials science, including ensemblelearning, unsupervised learning, active learning, and transferlearning. The directions for future research are proposed, including few-shot learning, and virtual sample generation.More importantly, we emphasize the significance of embedding material domain knowledge into ML and elaborate on thebasic idea for implementing this strategy.
文摘publishes only papers of the highest quality. All papers submitted are sent to referees familiar with the respective subjects for advice as to their eligibility for publication and needed improvement. The Editors make the final decision on publication in the light of such advice. Non-native English-speaking authors are advised to seek necessary linguistic assistance before submission of their manuscripts,although the Editors are prepared to help upgrade works of exceptional excellence. In addition to full length papers of up to 5000 words, the journal welcomes reviews, correspondence and shorter communications of less than 2500 words dealing with novel research techniques, analysis of current research, new concepts, etc.
