AI Design for High Entropy Alloys: Progress, Challenges and Future Prospects
Enzhi Xie, Chao Yang
Abstract
High-entropy alloys have demonstrated significant application potential in many industrial fields due to their outstanding comprehensive properties. However, the complex multi-component compositions pose challenges for traditional design approaches. In recent years, artificial intelligence (AI) technology, with its powerful capabilities in data analysis, prediction, and optimization, has provided new pathways for rapid discovery and performance modulation of high-entropy alloys. This paper systematically reviews the latest advancements in AI applications for high-entropy alloy design, covering key technologies such as machine learning models (e.g., active learning, generative models, transfer learning), high-throughput computing and experimental data processing, phase structure and property prediction. It also presents typical application cases, including compositional optimization, phase structure prediction, performance synergistic regulation, and novel material discovery. Although AI has significantly improved design efficiency and accuracy, challenges remain, such as the scarcity of high-quality data, insufficient model interpretability, and interdisciplinary integration. Future efforts should focus on building a more robust data ecosystem, enhancing model transparency, and strengthening closed-loop validation between AI and experimental science to advance intelligent design and engineering applications of high-entropy alloys.