Junyuan Hong

CSE PhD Student

Michigan State University

Biography

I am currently a final-year Ph.D. student of Computer Science and Engineering at ILLIDAN Lab@MSU, advised by Dr. Jiayu Zhou. Previously, I obtained my B.S. in Physics and M.S. in Computer Science at USTC. My research interests include Federated Learning and Data Privacy. Check my curricula vitae.

Interests

Federated Learning
Privacy
Manifold Learning

Education

MSc in Computer Science, 2018

University of Science and Technology of China
BSc in Physics, minor in CS., 2015

University of Science and Technology of China

Publications

Quickly discover relevant content by filtering publications.

Zhuangdi Zhu, Junyuan Hong, Steve Drew, Jiayu Zhou. Resilient and Communication Efficient Learning for Heterogeneous Federated Systems. ICML, 2022.

Junyuan Hong, Zhangyang Wang, Jiayu Zhou. Dynamic Privacy Budget Allocation Improves Data Efficiency of Differentially Private Gradient Descent. FAccT, 2022.

Junyuan Hong, Haotao Wang, Zhangyang Wang, Jiayu Zhou. Efficient Split-Mix Federated Learning for On-Demand and In-Situ Customization. ICLR, 2022.

Junyuan Hong, Zhuangdi Zhu, Shuyang Yu, Hiroko Dodge, Zhangyang Wang, Jiayu Zhou. Federated Adversarial Debiasing for Fair and Trasnferable Representations. KDD, 2021.

Junyuan Hong, Haotao Wang, Zhangyang Wang, Jiayu Zhou. Federated Robustness Propagation: Sharing Adversarial Robustness in Federated Learning. ArXiv, 2021.

Zhuangdi Zhu, Junyuan Hong, Jiayu Zhou. Data-Free Knowledge Distillation for Heterogeneous Federated Learning. ICML, 2021.

Junyuan Hong, Haotao Wang, Zhangyang Wang, Jiayu Zhou. Learning Model-Based Privacy Protection under Budget Constraints. AAAI, 2021.

Junyuan Hong, Yang Li, Huanhuan Chen. Variant Grassmann Manifolds: A Representation Augmentation Method for Action Recognition. TKDD, 2019.

Yang Li, Junyuan Hong, Huanhuan Chen. Short Sequence Classification Through Discriminable Linear Dynamical System. TNNLS, 2019.

Junyuan Hong, Huanhuan Chen, Feng Lin. Disturbance Grassmann Kernels for Subspace-Based Learning. KDD (Oral), 2018.

Yang Li, Junyuan Hong, Huanhuan Chen. Sequential Data Classification in the Space of Liquid State Machines. ECML, 2016.

Projects

Federated Learning

On the need of data privacy and more data, we strive to join the knowledge from a fair amount of users to train powerful deep neural networks without sharing data.

Differentially Private Learning

On the concern of data privacy, we aim to develop algorithms towards learning accurate models privately from data.

Subspace Learning

Supervised learning on subspace data which could model real data like skeleton motion.

Professional Activities

Internship

Research Intern with Dr. Lingjuan Lyu, Sony AI, 2022

Teaching Assistant

CSE 404: Introduction to Machine Learning, 2020 Fall

Award

Carl V. Page Memorial Graduate Fellowship, MSU, 2021
Student Travel Award, KDD, 2018
Outstanding Freshman Scholarship, USTC, 2015

Service

External Reviewer: ICML2022, KDD2022, WSDM2022, AISTATS2022, AAAI2022, AAAI2021, IJCAI2019, NeuroComputing, TKDD
Volunteer: Volunteer in KDD2018, KDD2021

Featured Publications

Junyuan Hong, Zhuangdi Zhu, Shuyang Yu, Hiroko Dodge, Zhangyang Wang, Jiayu Zhou

August 2021 KDD

Federated Adversarial Debiasing for Fair and Trasnferable Representations

A distributed domain/group debiasing framework for unsupervised domain adaptation or fairness enhancement.

Junyuan Hong, Huanhuan Chen, Feng Lin

June 2018 KDD (Oral)

Disturbance Grassmann Kernels for Subspace-Based Learning

In this paper, we focus on subspace-based learning problems, where data elements are linear subspaces instead of vectors. To handle this kind of data, Grassmann kernels were proposed to measure the space structure and used with classifiers, e.g., Support Vector Machines (SVMs). However, the existing discriminative algorithms mostly ignore the instability of subspaces, which would cause the classifiers to be misled by disturbed instances. Thus we propose considering all potential disturbances of subspaces in learning processes to obtain more robust classifiers.