Tongliang Liu

Tongliang Liu is an Associate Professor in the School of Computer Science at The University of Sydney. His research interests include computer vision, pattern recognition, deep learning, context learning, semi- and unsupervised learning, machine learning, data management and data science, computational imaging, and multimodal analysis and synthesis. He is also the Director of the Sydney AI Centre. Liu has authored and co-authored numerous research articles in top-tier conferences and journals, including ICML, NeurIPS, ICLR, CVPR, ICCV, and others. He has served in various editorial and review roles for prestigious conferences and journals.

Tongliang Liu is a Director at the Sydney AI Centre, University of Sydney, focusing on Machine Learning, Trustworthy Machine Learning, and foundation models. His research addresses challenges in noisy label learning, domain generalization, and robust model compression. Liu's work explores methods to improve model reliability and generalization, such as importance reweighting for noisy labels and conditional invariant adversarial networks for domain adaptation. He also investigates quantum machine learning and image quality assessment, emphasizing practical applications in real-world scenarios. His contributions highlight the importance of trustworthiness in AI systems, particularly in handling noisy data and ensuring model robustness.