My research interest is mainly on developing novel models and algorithms for tackling new challenges in deploying artificial intelligence systems to various real-world application domains. Regarding machine learning problems, I am mostly interested in the following topics: Low-resource learning: meta-learning, network pruning & quantization, deep generative models, self- & semi-supervised learning On-device learning: network compression (pruning, quantization, and knowledge distillation), continual learning, federated learning Safe and secure learning: uncertainty modeling & quantification, robustness to distributional shifts, defense against adversarial attacks Large-scale learning: meta-learning, neural architecture search, distributed & federated learning The application domains of interests include but are not limited to, visual recognition, natural language understanding, speech recognition, automatic drug/material discovery, healthcare and finance.