(National Engineering Research Center of Light Alloy Net Forming, School of Materials Science and Engineering, Shanghai Jiao Tong University，Shanghai 200240，China)
With the proposal of Material Genome Initiative and the rapid development of big data technology and artificial intelligence (AI), datadriven new material design has received widespread attention and gradually become an important method for researching and developing new materials. In recent years, researchers worldwide have carried out a lot of work on high-throughput calculations and AI-based materials design. They have obtained a large number of physical parameters that are difficult to obtain directly by experimental methods. That is the essence of data-driven materials design. Magnesium alloys have shown good application prospects in aerospace, electronic information and biomedical fields, but their low strength, poor plasticity, and low corrosion resistance have limited their further applications. This paper summarizes the recent researches for magnesium alloys based on first-principles calculation of density functional theory and the machine learning method of computer artificial intelligence, focusing on the mechanical properties, structures, microstructures of magnesium alloys(such as thermodynamic stability, energy barriers starting slip systems, relations between mechanical properties and constitutions/processes/microstructures), and corrosion resistance (such as the calculation of anode electrode potential, work function, cathode surface hydrolysis and hydrogen evolution reaction). Finally, the problems needed to be solved in the future are discussed.