The article provides a comprehensive overview of the recent advances and applications of machine learning in solid-state materials science. It begins by introducing the principles, algorithms, descriptors, and databases used in materials informatics. The authors then discuss various machine learning approaches for discovering stable materials, predicting their crystal structures, and calculating material properties. They also explore the use of machine learning in active learning and surrogate-based optimization for improving the rational design process. The interpretability and physical understanding gained from machine learning models are addressed, along with the challenges and future research directions in computational materials science. The article emphasizes the importance of large datasets, efficient algorithms, and high computing power in driving the success of machine learning in this field. It highlights the need for a standardized approach to data handling and the development of efficient implementations of machine learning methods for materials science.The article provides a comprehensive overview of the recent advances and applications of machine learning in solid-state materials science. It begins by introducing the principles, algorithms, descriptors, and databases used in materials informatics. The authors then discuss various machine learning approaches for discovering stable materials, predicting their crystal structures, and calculating material properties. They also explore the use of machine learning in active learning and surrogate-based optimization for improving the rational design process. The interpretability and physical understanding gained from machine learning models are addressed, along with the challenges and future research directions in computational materials science. The article emphasizes the importance of large datasets, efficient algorithms, and high computing power in driving the success of machine learning in this field. It highlights the need for a standardized approach to data handling and the development of efficient implementations of machine learning methods for materials science.