Kan Xu

Working Papers

Group-Sparse Matrix Factorization for Transfer Learning of Word Embeddings
Kan Xu, Xuanyi Zhao, Hamsa Bastani, Osbert Bastani
Major Revision, Management Science
- Preliminary version in ICML, 2021

Match Made with Matrix Completion: Efficient Offline and Online Learning in Matching Markets
Zhiyuan Tang, Wanning Chen, Kan Xu
First Place, POMS-HK Best Student Paper Award (Tang, 2025)

Optimal Multitask Linear Regression and Contextual Bandits under Sparse Heterogeneity
Xinmeng Huang, Kan Xu, Donghwan Lee, Hamed Hassani, Hamsa Bastani, Edgar Dobriban
Journal of the American Statistical Association, 2025

Multitask Learning and Bandits via Robust Statistics
Kan Xu, Hamsa Bastani
Management Science, 2025
Third Place, IBM Service Science Best Student Paper Award (2022)

Stochastic Bandits with ReLU Neural Networks
Kan Xu, Hamsa Bastani, Surbhi Goel, Osbert Bastani
International Conference on Machine Learning (ICML), 2024
[Code]

Uniformly Conservative Exploration in Reinforcement Learning
Wanqiao Xu, Jason Yecheng Ma, Kan Xu, Hamsa Bastani, Osbert Bastani
International Conference on Artificial Intelligence and Statistics (AISTATS), 2023

Group-Sparse Matrix Factorization for Transfer Learning of Word Embeddings
Kan Xu, Xuanyi Zhao, Hamsa Bastani, Osbert Bastani
International Conference on Machine Learning (ICML), 2021

Robust Generalization of Quadratic Neural Networks via Function Identification
Kan Xu, Hamsa Bastani, Osbert Bastani
ICML Uncertainty and Robustness in Deep Learning Workshop, 2021
ICML Overparameterization: Pitfalls & Opportunities Workshop, 2021

R Package: EbayesThresh
Kan Xu, Peter Carbonetto, Matthew Stephens, Bernard W. Silverman, Ludger Evers