How does Mean Absolute Error measure prediction errors in machine learning regression models? What makes MAE different from other evaluation metrics, such as Mean Squared Error? In which situations can MAE help developers understand model performance and prediction accuracy?