Fig. Overview of the computational data-driven framework for designing lightweight AlCoCrFeNi HEA with enhanced stiffness. (A)  To generate the dataset required for developing the ML model, 128 MD simulations of uniaxial tensile deformation are performed for the HEA configurations with random atomic composition. The dataset subsequently is utilized to develop computationally efficient GPR models. Large-scale predictions of the desired responses (Young's modulus (Y) and density (ρ)) for unknown HEA configurations are produced by using the generalized GPR model. The ML models are utilized further for performing sensitivity analysis and genetic algorithm based multi-objective optimization to determine the optimum atomic composition of AlCoCrFeNi HEA by minimizing the density and maximizing the stiffness. (B) Explainability analysis of the constructed ML models, leading to enhanced insights. 

Fig. Machine learning-based computational framework coupled with molecular dynamics simulations. The seamless integration of machine learning and MD simulation is exploited here to unravel the deep insights concerning the atomic composition of AlCoCrFeNi HEA for achieving optimal ballistic performance. (A) Sobol sequence sampling-based typical random configurations of AlCoCrFeNi HEA modelled in LAMMPS environment (B) Flow diagram of the complete computational framework adopted in the present study (C) Post-impact atomistic deformation mechanism of HEA configurations suggested by large-scale physics-based investigation. 

Fig. Machine learning based computational insights for the critical mechanical characteristics of graphene. Machine learning (ML) assisted framework for efficient prediction of mechanical properties. The external and internal features which influence the mechanical properties of graphene are considered as sparse input parameters. The D-optimal algorithm is exploited to create a sample space based on the set of input features, followed by molecular dynamics simulations for training the ML model, and finally formation of the predictive ML model.