<p>Residual stresses, which remain within a component after processing, can deteriorate performance. Accurately determining their full-field distributions is essential for optimizing the structural integrity and longevity. However, the experimental effort required for full-field characterization is impractical. Given these challenges, this work proposes a machine learning (ML) based Residual Stress Generator (RSG) to infer full-field stresses from limited measurements. The RSG employs a U-Net architecture, a convolutional neural network originally developed for image segmentation, to generate complete residual stress distributions by learning the latent structure of stress fields from mechanics-based modeling. An extensive dataset was initially constructed by performing numerous simulations of the residual stress-inducing process, friction stir processing, with a diverse parameter set. The ML model was trained through systematic hyperparameter tuning to capture spatial features with high fidelity. Then, the model’s ability to generate simulated stresses was evaluated, and it was ultimately tested on actual characterization data to validate its effectiveness. The model’s prediction of simulated stresses demonstrates excellent predictive accuracy and a significant degree of generalization, showcasing the transformative potential of AI in reconstructing stress distributions. The RSG’s performance in predicting experimentally characterized data highlights the feasibility of the proposed approach in providing a comprehensive understanding of residual stress distributions from limited measurements, thereby significantly reducing experimental efforts.</p>