Paper Title
Exchange Bias Prediction for AFM-FM Core Shell Nanoparticles and Thin Films Using Machine Learning

Abstract
Exchange bias (EB) plays a crucial role in stabilizing and tuning the magnetic behavior of materials for applications in spintronics, magnetic storage, sensors, and energy-efficient devices. In particular, antiferromagnetic - ferromagnetic (AFM/FM) core–shell nanoparticles and thin film heterostructures exhibit enhanced EB effects due to interfacial exchange coupling, leading to improved coercivity,enhanced Neel temperature and tunable blocking temperatures. However, accurately predicting EB in such complex systems using conventional theoretical and computational approaches remains challenging due to high computational cost and the intricate dependence on structural, compositional, and interfacial parameters. In this study, we develop a machine learning (ML)-based framework to predict the exchange bias field in AF/FM core–shell nanoparticles and thin films, providing a fast and efficient alternative to traditional methods. Multiple ML models, including ensemble and boosting techniques, are systematically evaluated to capture the nonlinear relationships governing EB behavior. Comparative analysis also highlights the strengths and limitations of different algorithms in capturing extreme EB values and complex interfacial effects. Keywords - Exchange bias, Core Shell Nanoparticles, Thin Films, Machine Learning, AFM-FM systems