Class Sample

Class sample research focuses on improving the performance of machine learning models in scenarios where the test data may contain unseen or unknown classes, a departure from traditional closed-set learning. Current research emphasizes developing robust methods for identifying and handling these unknown classes, often employing techniques like active learning, self-supervised learning, and generative models within various architectures including deep neural networks and mixture-of-experts models. This work is crucial for building more reliable and safe AI systems in real-world applications where encountering novel data is inevitable, such as in autonomous driving, medical diagnosis, and anomaly detection.