机器学习在二维材料探索中的应用

Application of machine learning in two-dimensional material exploration

当前,机器学习已经成为探索和拓展二维材料家族的重要研究手段。传统实验与计算方法在研究二维材料时容错率低,并且需要花费大量时间和研发成本。机器学习因为拥有强大的数据处理能力和灵活多样的算法模型,能够绕过求解理论计算复杂的泛函方程以及缓慢的实验过程,缩短研究周期帮助减少发现和理解二维材料的时间和成本,以数据为基础高效预测扩展二维材料体系并探究其实验合成以及应用的潜力。将围绕机器学习的方法、机器学习在二维材料设计与合成、机器学习在二维材料物性与应用的探索等方面,详细介绍相关的前沿进展,最后对机器学习在二维材料领域开展研究所面临的挑战与发展趋势进行了展望。

At present,machine learning has become an important research method to expand and explore the family of two-dimensional(2 D)materials.In the study of 2 D materials,the traditional experiment and calculation methods have low fault-tolerance and need a lot of time and cost.Machine learning,because of its powerful data processing ability and flexible algorithm models,can bypass the solution of complex functional equations and slow experimental process,and shorten the research period.Using machine learning to explore 2 D materials can reduce the time and cost of discovering and understanding 2 D materials,efficiently predict and expand 2 D material systems based on data,and explore their potential in experimental synthesis and application.The methods of machine learning,machine learning in the design and synthesis of 2 D materials,machine learning in the exploration of 2 D material properties and applications,etc.were reviewed,and the relevant cutting-edge progress was introduced in detail.Finally,the challenges and development trends of machine learning in the field of 2 D materials were prospected.