Novel Class Discovery

Novel Class Discovery (NCD) is a machine learning technique aimed at identifying new categories in unlabeled data by leveraging prior knowledge of known categories. This task divides the data into two sets: a set of labeled known categories and a set of unlabeled unknown categories. The core objective of NCD is to enhance the model's ability to discover new categories, thereby improving the system's generalization performance and adaptability. It has significant application value, especially in continuous learning and classification tasks in dynamic environments.