The process of using a trained machine learning model to make predictions or decisions on new, previously unseen data. Inference involves feeding input data through the trained model to generate outputs, typically requiring less computational resources than training phases.
