Researchedit

This page summarizes the main research directions in Qiao Xinbao's academic wiki. It functions as a compiled map of linked topic pages rather than a static list of interests. The current center of gravity is AI and networks.

Research thesisedit

The common thread is data-process reliability under networked constraints: how to design learning algorithms when data are selected, removed, synthesized, siloed, biased, privacy-constrained, or communication-constrained. The work is primarily positioned in AI and networks, machine unlearning, synthetic data, and data centric ML. Following the wiki pattern, each concept page records a stable local synthesis and points outward to papers, institutions, and adjacent methods, so later updates can revise the map rather than restart from raw notes.

AI and networksedit

AI and Networks covers the intersection of AI with networking and communication systems: AI for communication, communication for AI, decentralized learning, data pruning, and collaborative evaluation. In the current CUHK doctoral stage, this line includes distributed computing for Wasserstein barycenters, where multiple local distributions are combined into a shared distributional reference without treating raw-data pooling as the default assumption.

Machine unlearningedit

Machine Unlearning studies certified data removal and low-cost update mechanisms after deletion requests. Related pages include Hessian-Free Online Certified Unlearning, Beyond Binary Erasure: Soft-Weighted Unlearning for Fairness and Robustness, DynFrs: An Efficient Framework for Machine Unlearning in Random Forest, Influence Functions, and Certified Data Removal.

Synthetic dataedit

Synthetic Data studies recursive synthetic-data training, Data Selection, Sample Selection Bias, Model Collapse, and collaborative mitigation in Data Silos. The central paper is When Sample Selection Bias Precipitates Model Collapse.

Data centric ML and trustworthy AIedit

Data Centric ML covers data selection, valuation, filtering, and evaluation. Trustworthy AI connects unlearning, fairness, robustness, privacy, security, interpretability, and reliability.

Geometry and distributed learningedit

Wasserstein Geometry, Distributed Wasserstein Barycenter, and Distributed Learning provide tools for collaborative evaluation, optimal-transport proxies, decentralized data access, and distributional references for networked AI systems.