Junxue ZHANG is a Ph.D. student of computer scinece at iSINGLab, Hong Kong Univeristy of Science and Technology. He is under supervision of Prof. Kai CHEN. His reseach interests are data center networking, AI systems and privacy preserving computation. He is the co-founder/CTO of Clustar Technology Co., Ltd. He is in the TSC of federated machine learning framework: FATE.

Before joining HKUST, he received his BSc and MSc from Southeast Univeristy.

Research Projects

Data Center Networking

1. Traditional DCN: Load balance in the wild [Hermes, SIGCOMM 17] , ECN with RTT variations [ECN#, CoNEXT 19]

2. Machine Learning for Networking: Efficient neural network deployment in kernel datapath [LiteFlow, SIGCOMM 22 (To appear)] , Multi-objective RL for CC [MOCC, EuroSys 22]

3. Networking for Machineg Learning: New transport [MLT, APNet 19] , Efficient parameter aggregation [RAT, APNet 20]

Privacy-preserving Computation

1. Federated Learning: Measurement [Quantifying, FL-IJCAI 2019] , Verification [Aegis, FL-IJCAI 2021] , Hardware acceleration [FLASH, under submission] , Communication acceleration [undergoing]

2. Fully Homomorphic Encryption: Hardware acceleration [undergoing]

3. New Secure Protocol Design: HE + DP [Sphinx, S&P22]

Publications

Conference Paper

1. LiteFlow: Towards High-performance Adaptive Neural Networks for Kernel Datapath
   Junxue Zhang, Chaoliang Zeng, Hong Zhang, Shuihai Hu, Kai Chen
   SIGCOMM 2022 (To appear)

2. Multi-Objective Congestion Control
   Yiqing Ma, Han Tian, Xudong Liao, Junxue Zhang, Weiyan Wang, Kai Chen, Xin Jin
   EuroSys 2022
   [paper]

3. Sphinx: Enabling Privacy-Preserving Online Learning over the Cloud
   Han Tian, Chaoliang Zeng, Zhenhang Ren, Di Chai, Junxue Zhang, Kai Chen, Qiang Yang
   S&P 2022
   [paper]

4. Enabling ECN for Datacenter Networks with RTT Variations
   Junxue Zhang, Wei Bai, Kai Chen
   CoNEXT 2019
   [paper] [code]

5. Resilient Datacenter Load Balancing in the Wilds
   Hong Zhang, Junxue Zhang, Wei Bai, Kai Chen, Chowdhury Mosharaf
   SIGCOMM 2017
   [paper] [code]

Workshop Paper

1. Aegis: A Trusted, Automatic and Accurate Verification Framework for Vertical Federated Learning
   Cengguang Zhang, Junxue Zhang, Di Chai, Kai Chen
   FL-IJCAI 2021 (Best Application Paper Award)
   [paper]

2. Resilient Allreduce Tree for Distributed Machine Learning
   Xinchen Wan, Hong Zhang, Hao Wang, Shuihai Hu, Junxue Zhang, Kai Chen
   APNet 2020
   [paper]

3. Quantifying the Performance of Federated Transfer Learning
   Qinghe Jing, Weiyang Wang, Junxue Zhang, Han Tian, Kai Chen
   FL-IJCAI 2019 (Best Student Paper Award)
   [paper]

4. Rethinking Transport Layer Design for Distributed Machine Learning
   Jiachen Xia, Gaoxiong Zeng, Junxue Zhang, Weiyang Wang, Wei Bai, Junchen Jiang, Kai Chen
   APNet 2019
   [paper]

5. Bridging the Edge-Cloud Barrier for Real-time Advanced Vision Analytics
   Yiding Wang, Weiyang Wang, Junxue Zhang, Junchen Jiang, Kai Chen
   HotCloud 2019
   [paper]

Google Scholar

Industry Experience

2018 - Now, Co-founder/CTO, Clustar Technology Co., Ltd

Professional Services & Activities

1. FL-IJCAI 2021, FL-IJCAI 2022, FL-AAAI 2022 TPC Member

2. FATE Maintainer

Awards

1. SIGCOMM 17 Travel Grants, 2017

2. Postgraduate Fellowship, HKUST

3. Google Excellent Student Scholarship, 2012

4. National Scholarship 2014, 2011

Miscellaneous

Junxue is an amateur photographer mainly at landscape and sometimes portrait. He likes coffee, and is a fresh man in latte art. He also has a not very large collection of Lego. Junxue has 1 cat named Little Silver (小盐) & 2 dogs named Little White (小白) and Big White (大白).