Xinwen Zhang
Ph.D. Student in Computer & Information Science, Temple University
I am a Ph.D. student in Computer & Information Science at Temple University, advised by Dr. Hongchang Gao and also working closely with Dr. Richard Souvenir. My research focuses on designing large-scale optimization algorithms, including:
- Compositional optimization for AUC maximization
- Optimization beyond minimization (e.g., minimax, bilevel, multi-level, etc)
- Optimization for large language models (pre-training and post-training)
Warning
Problem: The current name of your GitHub Pages repository ("Solution: Please consider renaming the repository to "
http://".
However, if the current repository name is intended, you can ignore this message by removing "{% include widgets/debug_repo_name.html %}" in index.html.
Action required
Problem: The current root path of this site is "baseurl ("_config.yml.
Solution: Please set the
baseurl in _config.yml to "Education
-
Temple UniversityPh.D. Student in Computer & Information Science
Advisor: Dr. Hongchang GaoAug. 2022 - present -
Tongji UniversityB.S. in MathematicsSep. 2017 - Jul. 2021
Honors & Awards
-
Outstanding Graduate Research Assistant Award in CIS at Temple University, ICML Gold Reviewer, ICML Financial Aid Award.2026
-
Female Student Award in Data Science at ICDM, ICDM Student Travel Award.2025
-
ICML Travel Award.2024
-
KDD Student Travel Award, NeurIPS Scholar Award, WiML Travel Funding.2023
News
08/2026: 📄 One paper on Federated AUC Maximization was accepted to ICDM 2026.
05/2026: 📄 Three papers on Distributed Optimization were accepted by ICML 2026.
04/2026: 🏅 Been awarded the Outstanding Graduate Research Assistant in CIS at Temple University!
02/2026: 📄 One paper on Quantum Networks was accepted to IEEE Transactions on Networking 2026.
11/2025: 🏅 Been awarded the Female Student Award in Data Science at ICDM 2025!
11/2025: 📘 Delivered a tutorial on Federated Optimization at ICDM 2025.
09/2025: 📄 One paper on Compositional AUC Maximization was accepted to NeurIPS 2025.
09/2025: 📄 One paper on Sharpness-Aware AUC Maximization was accepted to ICDM 2025.
08/2025: 📘 Delivered a tutorial on Federated Optimization at IJCAI 2025.
04/2025: 📄 One paper on Quantum Networks was accepted to IWQoS 2025.
05/2024: 📄 One paper on Federated Compositional AUC Maximization was accepted to ICML 2024.
09/2023: 🖇️ One poster on Federated Compositional AUC Maximization was accepted to the WiML 2023 Workshop.
09/2023: 📄 One paper on Federated Compositional AUC Maximization was accepted to NeurIPS 2023.
08/2023: 📘 Delivered a tutorial on Distributed Optimization at KDD 2023.
Publications
2026
Faster Federated Smoothed Minimax Optimization
Xinwen Zhang,
Liang Zhan,
Hongchang Gao
In Proceedings of the 26th IEEE International Conference on Data Mining (ICDM 2026).
LS2MC-GDA: A Smoothed Algorithm for Federated Stochastic Multi-level Compositional Minimax Optimization [Paper]
Xinwen Zhang,
Richard Souvenir,
Hongchang Gao
In Proceedings of the 43rd International Conference on Machine Learning (ICML 2026).
Distributed Stochastic K-Level Optimization Over Networks [Paper]
Xinwen Zhang,
Yihan Zhang,
Hongchang Gao,
Heng Huang
In Proceedings of the 43rd International Conference on Machine Learning (ICML 2026).
Convergence Analysis of Decentralized Hessian-/Jacobian-Free Algorithm for Nonconvex Stochastic Bilevel Optimization [Paper]
Yihan Zhang,
Xinwen Zhang,
My T. Thai,
Jie Wu,
Hongchang Gao
In Proceedings of the 43rd International Conference on Machine Learning (ICML 2026).
Swapping and Purification Scheme Optimization for Entanglement Distribution in Quantum Networks [Paper]
Jiyao Liu,
Xinwen Zhang,
Xinliang Wei,
Xuanzhang Liu,
Yuzhou Chen,
Hongchang Gao,
Yu Wang
In IEEE Transactions on Networking (ToN 2026).
2025
On the Convergence of Stochastic Smoothed Multi-Level Compositional Gradient Descent Ascent [Paper]
Xinwen Zhang,
Hongchang Gao
In Proceedings of the 39th Conference on Neural Information Processing Systems (NeurIPS 2025).
Sharpness-Aware Optimization Through Variance Suppression on Deep AUC Maximization [Paper]
Xinwen Zhang,
Hongchang Gao
In Proceedings of the 25th IEEE International Conference on Data Mining (ICDM 2025).
Joint Swapping and Purification with Failures for Entanglement Distribution in Quantum Networks [Paper]
Jiyao Liu,
Xinwen Zhang,
Xinliang Wei,
Xuanzhang Liu,
Yuzhou Chen,
Hongchang Gao,
Yu Wang
In IEEE/ACM 33rd International Symposium on Quality of Service (IWQoS 2025).
2024
A Federated Stochastic Multi-level Compositional Minimax Algorithm for Deep AUC Maximization [Paper]
Xinwen Zhang,
Ali Payani,
Myungjin Lee,
Richard Souvenir,
Hongchang Gao
In Proceedings of the 41st International Conference on Machine Learning (ICML 2024).
2023
Federated Compositional Deep AUC Maximization [Paper]
Xinwen Zhang,
Yihan Zhang,
Tianbao Yang,
Richard Souvenir,
Hongchang Gao
In Proceedings of the 37th Conference on Neural Information Processing Systems (NeurIPS 2023).
Preprint
Federated Stochastic Minimax Optimization under Heavy-Tailed Noises [Paper]
Xinwen Zhang,
Hongchang Gao
arXiv preprint arXiv:2511.04456.
On Provable Benefits of Muon in Federated Learning [Paper]
Xinwen Zhang,
Hongchang Gao
arXiv preprint arXiv:2510.03866.
Nonconvex Decentralized Stochastic Bilevel Optimization under Heavy-Tailed Noises [Paper]
Xinwen Zhang,
Yihan Zhang,
Heng Liang,
Hongchang Gao
arXiv preprint arXiv:2509.15543.
Tutorial & Workshop
2025
Federated Stochastic Compositional and Bilevel Optimization [Link]
Hongchang Gao,
Xinwen Zhang
In Proceedings of the 25th IEEE International Conference on Data Mining (ICDM 2025).
Federated Stochastic Compositional and Bilevel Optimization [Link]
Hongchang Gao,
Xinwen Zhang
In Proceedings of the 34th International Joint Conference on Artificial Intelligence (IJCAI 2025).
2023
A Fast Federated Stochastic Compositional Minimax Optimization Algorithm with Variance Reduction
Xinwen Zhang,
Richard Souvenir,
Hongchang Gao
In the 18th Women in Machine Learning Workshop (WiML 2023).
Distributed Optimization for Big Data Analytics: Beyond Minimization [Link]
Hongchang Gao,
Xinwen Zhang
In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023).
Service
Reviewer
- ICLR, AISTATS, AAAI, ICML, KDD, NeurIPS (2026)
- ICLR, AISTATS, AAAI, ICML, KDD, NeurIPS (2025)
- AAAI, ICML, NeurIPS (2024)
- Journal of Combinatorial Optimization, Neural Networks, Journal of Optimization Theory and Applications.
Teaching
- CIS3715 Principles of Data Science — Spring 2023, Spring 2024
- CIS1051 Introduction to Problem Solving and Programming in Python — Fall 2022, Spring 2023, Fall 2023, Spring2024, Fall 2024