I am ✨ Xinbao Qiao (Sheen-bao Ciao /ʃɪnbaʊ/ /tʃjaʊ/), 🐼 Namesake of Panda Xinbao, a final-year master student in the Zhejiang University.
I graduated from Shandong University with a bachelor’s degree. Now I study at ZJUI with a master’s degree, advised by Meng Zhang.
My research interest focus on building data-centric ML systems that ensure trustworthy AI systems, and deploy these systems to solve real-world challenges, e.g., distributed systems and healthcare/biomedicine.
Publications
Beyond Binary Erasure: Soft-Weighted Unlearning for Fairness and Robustness
Xinbao Qiao, Ningning Ding, Yushi Cheng, Meng Zhang
- Corrective Machine Unlearning, Influence Function, Fairness, Robustness
TL;DR: Addressing the challenge of excessive unlearning in non-privacy-focused scenarios through soft-weighted strategies.
Hessian-Free Online Certified Unlearning
Xinbao Qiao, Meng Zhang, Ming Tang, Ermin Wei
- Approximate Machine Unlearning, Certified Data Removal, Privacy
TL;DR: addressing second-order certified unlearning gaps in non-convex settings and nonconvergence conditions, while achieving the most efficient data removal.
DynFrs: An Efficient Framework for Machine Unlearning in Random Forest
Shurong Wang, Zhuoyang Shen, Xinbao Qiao, Tongning Zhang, Meng Zhang
- Exact Machine Unlearning, Random Forest, Privacy
TL;DR: combining subsampling methods, the lazy tag strategy, and the robustness of Extremely Randomized Trees to achieve Random Forest unlearning in dynamic environments.
Educations
- 2022.09 - 2025.06, Master, Artificial Intelligence, Zhejiang University
- 2018.09 - 2022.07, Bachelor, Communication Engineering, Shandong Univeristy.