معرفی
Junfeng Wen is an Assistant Professor at the School of Computer Science, Carleton University. He holds a Ph.D. in Statistical Machine Learning from the University of Alberta, where he worked with Prof. Dale Schuurmans and Prof. Russ Greiner. His research focuses on foundational machine learning algorithms, including reinforcement learning, transfer learning, federated learning, and domain adaptation. He teaches courses such as Reinforcement Learning and Machine Learning at Carleton University.
Education:
- Ph.D. in Statistical Machine Learning (2013–2020), University of Alberta
- M.Sc. in Computing Science (2011–2013), University of Alberta
- Bachelor of Engineering (with Honors, 2007–2011), College of Computer Science and Technology, CKC College, Zhejiang University
Research Interests:
Dr. Wen’s work emphasizes advancing theoretical understanding and practical applications of machine learning. Key areas include federated learning frameworks for privacy preservation, reinforcement learning for decision-making systems, and multi-source domain adaptation techniques. His research bridges algorithmic innovation with real-world challenges in distributed systems and biological modeling.
Publications:
His recent articles explore federated learning optimization, decentralized collaboration strategies, and reinforcement learning applications in behavioral studies. Notable contributions include domain aggregation networks for cross-domain adaptation and proxy model-sharing techniques in decentralized learning. The work spans theoretical advancements and practical implementations, often addressing scalability and privacy concerns in distributed environments.
Professional Contributions:
- Conference reviewer for NeurIPS, ICML, ICLR, AAAI, IJCAI, and ACML
- Journal reviewer for JMLR, Machine Learning Journal (Editorial Board member), and TPAMI
Labs and Projects:
He contributes to open-source projects like Domain Aggregation Networks (DARN) for multi-source domain adaptation and Frank-Wolfe for Covariate Shift. These repositories reflect his commitment to reproducible research and community engagement.
