Computation and machine learning for materials: Past, present, and future perspectives

S Alosious and M Jiang and TF Luo, MRS BULLETIN, 50, 1212-1224 (2025).

DOI: 10.1557/s43577-025-00959-y

Computational methods and machine learning (ML) are reshaping materials science by accelerating their discovery, design, and optimization. Traditional approaches such as density functional theory and molecular dynamics have been instrumental in studying materials at the atomic level. However, their high computational cost and, in certain cases, limited accuracy can restrict the scope of in silico exploration. ML promises to accelerate material property prediction and design. However, in many areas, the volume and fidelity of the data are critical barriers. Active learning can reduce the reliance on large data sets, and simulation has emerged as a critical tool for generating data on the fly. Despite these advances, challenges remain, particularly in data quality, model interpretability, and bridging the gap between computational predictions and experimental validation. Future research should develop automated frameworks capable of designing and testing materials for specific applications, and integrating ML with traditional simulations and experiments can contribute to this goal.Graphic abstractA schematic diagram illustrates the evolution of computational materials science, from early density functional theory (DFT) and molecular dynamics (MD) methods to the integration of high-throughput simulations and machine learning (ML). It also highlights artificial intelligence (AI)-driven discovery, generative models, and quantum computing as key future directions for accelerating materials innovation

Return to Publications page