Yajie Bao

PhD candidate, Statistics School of Mathematical Sciences Shanghai Jiao Tong University Advisor: Dong Han Email: baoyajie2019stat@sjtu.edu.cn Google scholar

About Me

I am currently a final-year Ph.D. student in Statistics at Shanghai Jiao Tong University. Before coming to SJTU, I received my B.S. degree in Statistics from East China Normal University. I have been working with Mingrui Liu on federated learning and deep learning theory since November 2021. At SJTU, I'm a member of Haojie Ren's research group on conformal inference. I was fortunate to visit Prof. Changliang Zou's group at Nankai University from April-August, 2023.

Research interests

• Comformal inference

• Deep learning theory

• Federated learning

Recent papers

• Provable Benefits of Local Steps in Heterogeneous Federated Learning for Neural Networks: A Feature Learning Perspective
  Yajie Bao, Michael Crawshaw and Mingrui Liu.
  International Conference on Machine Learning (ICML), 2024.

• CAP: A General Algorithm for Online Selective Conformal Prediction with FCR Control
  Yajie Bao, Yuyang Huo, Haojie Ren and Changliang Zou.
  arxiv, 2024+.

• Selective Conformal Inference with False Coverage-statement Rate Control
  Yajie Bao, Yuyang Huo, Haojie Ren and Changliang Zou.
  Biometrika, 2024.

• Global Convergence Analysis of Local SGD for Two-layer Neural Network without Overparameterization
  Yajie Bao, Amarda Shehu, Mingrui Liu.
  Advances in Neural Information Processing Systems (NeurIPS), 2023.

• Federated Learning with Client Subsampling, Data Heterogeneity, and Unbounded Smoothness: A New Algorithm and Lower Bounds
  Michael Crawshaw, Yajie Bao and Mingrui Liu.
  Advances in Neural Information Processing Systems (NeurIPS), 2023.

Service

Reviewer: ICML 2022, 2024; NeurIPS 2022, 2023; ICLR 2024; AISTATS 2024.