Weijia Zhang

Dr. Weijia Zhang is a Lecturer in Data Science and Applied Statistics at the University of Newcastle (Australia), focusing on causal inference and machine learning from ambiguous data. He holds a PhD from the University of South Australia (2018) and has held roles as Associate Professor at Southeast University (China) and Research Fellow at the University of South Australia. His research bridges causal inference and machine learning, addressing challenges in weakly supervised learning (e.g., multi-instance partial-label learning) and survival analysis under dependent censoring.

• Affiliations: University of Newcastle (current), Southeast University (2021–2023), University of South Australia (2018–2020)
• Teaching: Courses on predictive analytics, unsupervised methods, statistical decision-making, and time series forecasting

Research interests include causal representation learning, weakly supervised learning from noisy/ambiguous labels, and applications in healthcare, telecommunications, and automotive industries. Key projects include developing causal methods for battery lifespan prediction with Chang'an Automobile and models for dependent censoring in survival analysis.

Publications (over 30+): Focus on causal inference, weak supervision, and survival analysis. Notable contributions include CausalPLL+, MiplGp, and TEDVAE algorithms. Editor for Data Science and Engineering and reviewer for top conferences (NeurIPS, ICML, AAAI).

• Grants: $125k from Changan Automobile, $63.5k from National Natural Science Foundation of China, and institutional funding ($132.5k total)
• Awards: Top Reviewer Awards at AAAI (2021) and UAI (2022)

Labs/Teams: Leads research on causal representation learning and survival analysis within the School of Information and Physical Sciences, collaborating internationally with institutions in China, Switzerland, and the US.