Data-Driven Security
In this area, we study security and privacy issues driven by massive data, including system data, network data, application data, etc. The commonly employed techniques include Machine Learning (ML), Natural Language Processing (NLP), graph theory, as well as other related techniques. Some previous and ongoing research includes: (1) Cyber Threat/Crime Mining and Analysis; (2) Anonymization and De-anonymization; (3) Medical Application/Data Security and Privacy; (4) Multimedia Security; and (5) Password and CAPTCHA Security.

AI and Security
In this area, we study security and privacy issues from two perspectives. On one hand, we study how to employ AI techniques to improve system, network, and application security and privacy. On the other hand, we study the security and privacy of AI techniques/systems themselves. Some previous and ongoing research includes: (1) Adversarial Learning, in domains of image, audio, text, etc.; (2) Machine Learning (include Deep / Federated / Distributed Learning) Security and Privacy; (3) Verifiable and Certifiable Machine Learning; (4) Deep Learning Interpretability and Model Robustness Quantification; and (5) Secure Multi-Party Computing and Learning.

Software and System Security
In this area, we study the security issues of various software and systems. Some previous and ongoing research includes: (1) Smart Fuzzing; (2) AI-aided Program Analysis; (3) ICS Security; (4) Blockchain; and (5) Vulnerability Mining and Analysis.

Big Data Mining and Analysis
In this area, we focus on mining knowledge from massive data and understanding the evolution and characteristics of interested systems. Some previous and ongoing research includes:  (1) Social Network Computing and Analytics and (2) Graph and Network Embedding.