A new framework called Learning-Augmented Power System Operations (LAPSO) has been proposed to improve the efficiency and reliability of power system operations. LAPSO treats machine learning as an explicit component of decision-making, allowing for more accurate predictions and optimization. The framework provides a unified view of optimization and machine learning, addressing the gap between standalone ML pipelines and downstream optimization problems. It includes mathematical templates for generalizing decision-independent predictors and learned surrogates that enter optimization as auxiliary constraints. LAPSO is demonstrated on stability-constrained optimization and objective-based forecasting, showing improved performance compared to traditional methods. --- Why it matters: This matters because power system operations are becoming increasingly complex due to the integration of renewable energy sources and inverter-based resources. A unified framework like LAPSO can help optimize decision-making and improve economic efficiency, security, and robustness. Source: https://arxiv.org/abs/2505.05203