My research interests are broadly in
Machine Learning Theory: Understanding the fundamental principles of machine learning, e.g., network capacity, optimization method and generalization ability, and providing insights for algorithms.

Self-Supervised Learning: Self-supervised learning aims to pre-train a powerful feature extractor through a self-supervised task using a large amount of unlabeled data, such that the learned image representations can be efficiently adapted to downstream tasks.

Few-Shot Learning: Few-shot learning is the problem of making predictions on new tasks containing only a few samples with supervised information.