Ga Wu
About
Dr. Ga Wu’s research centers on machine learning, with a particular emphasis on self-supervised and cross-domain contrastive representation learning to advance multimodality AI. His work extends to model explanation and model editing, where understanding the relationships among model parameters, input–output behavior, and learned latent representations is essential. Recently, Dr. Wu has focused on two key directions: (1) representation learning and generation from 3D point clouds and meshes, and (2) minimally invasive machine unlearning through optimization-free, inference-time representation augmentation. He leads the Applied Machine Learning Research Lab (DAMLR), which actively collaborates with numerous industrial partners and research institutes to push the boundaries of theoretical machine learning, while offering students hands-on opportunities across more than ten active industry collaborations.
Note to prospective students
The Applied Machine Learning Research Lab (DAMLR) is seeking self-motivated students with a strong interest in advancing representation learning. We value candidates who not only apply AI techniques but strive to understand the deep mathematical principles behind them. A solid background in math is therefore essential, along with strong coding ability—ideally demonstrated through consistent contributions to a GitHub repository over the past one to two years. For MSc applicants, funding is not guaranteed at admission; rather, it is offered once a student clearly demonstrates research potential. Those who show strong capability will gain not only funding but also opportunities to collaborate with our industry partners. We value students who embrace challenge and deliver results efficiently—not simply those who say they “tried their best” but mess things up.