Xijun Li (李希君)

alt text 

Assistant Professor @ Shanghai Jiao Tong University (SJTU)
Ph.D. @ University of Science and Technology of China (USTC)

E-mail: lixijun@sjtu.edu.cn (preferred), xijun.lee@hotmail.com
Address: Room 1411, Software Building, SJTU, Minhang District, Shanghai, China

[Homepage@SJTU] [Google Scholar] [Curriculum Vitae] [Github] [Countdown]

Messages for Collaborators
  • For prospective students: We are looking for self-motivated PhD, Master and Undergraduate students who have a strong interest in Learning to Optimize and Large Language Model for Optimization/Reasoning. If you want to work with me, please feel free to drop me an email lixijun@sjtu.edu.cn along with your CV to arrange an appointment.

  • For research collaborators: I welcome students and researchers to initiate academic collaborations with me. You can schedule a discussion meeting with me through calendar tool.

News

About

Currently, Xijun is an assistant professor in Shanghai Jiao Tong University and a faculty member of the Shanghai Key Labraotry of Scalable Computing and Systems. Previously, he has served as a princinpal researcher in Huawei Noah’s Ark Lab between 2018 and 2024. He has received his Ph.D. degree from the University of Science and Technology of China in March 2024 (HUAWEI-USTC Joint Ph.D. Program) under the supervision of Prof. Jie Wang and received the master degree under the supervision of Prof. Jianguo Yao from Shanghai Jiao Tong University in 2018. He has published many papers on top peer-reviewed conferences and journals (such as TPAMI, NeurIPS, ICLR, ICML, KDD, ICDE, SIGMOD, DAC, etc.) and applied/published a series of patents with Huawei Noah's Ark Lab. He has also won the championship of student learderboard in Dual Track of NeurIPS’21 ML4CO Competition. His recent research interests focus on Learning to Optimization (L2O) and Large Language Model for Optimization/Reasoning. Last but not least, he is one of core developers of the Huawei Cloud OptVerse AI Solver and Huawei Pangu Large Language Models.

Research

Recent Interests

Publications

  1. Haoyang Liu, Jie Wang, Wanbo Zhang, Zijie Geng, Yufei Kuang, Xijun Li, Bin Li, Yongdong Zhang, Feng Wu: MILP-StuDio: MILP Instance Generation via Block Structure Decomposition. NeurIPS 2024 [pdf]

  2. Jie Wang, Zhihai Wang, Xijun Li, Yufei Kuang, Zhihao Shi, Fangzhou Zhu, Mingxuan Yuan, Jia Zeng, Yongdong Zhang, Feng Wu: Learning to Cut via Hierarchical Sequence/Set Model for Efficient Mixed-Integer Programming. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) [pdf]

  3. Zhihai Wang, Lei Chen, Jie Wang, Yinqi Bai, Xing Li, Xijun Li, Mingxuan Yuan, Jianye HAO, Yongdong Zhang, Feng Wu: A Circuit Domain Generalization Framework for Efficient Logic Synthesis in Chip Design. ICML 2024 [pdf]

  4. Yufei Kuang, Jie Wang, Yuyan Zhou, Xijun Li, Fangzhou Zhu, Jianye HAO, Feng Wu: Towards General Algorithm Discovery for Combinatorial Optimization: Learning Symbolic Branching Policy from Bipartite Graph. ICML 2024 [pdf]

  5. Hankun Cao, Shai Bergman, Si Sun, Yanbo Albert Zhou, Xijun Li, Jun Gao, Zhuo Cheng, Ji Zhang: Answering the Call to ARMs with PACER: Power-Efficiency in Storage Servers. MSST 2024 [pdf]

  6. Kecheng Huang, Xijun Li, Mingxuan Yuan, Ji Zhang, Zili Shao: Joint Directory, File and IO Trace Feature Extraction and Feature-based Trace Regeneration for Enterprise Storage Systems. ICDE 2024 [pdf]

  7. Yufei Kuang, Jie Wang, Haoyang Liu, Fangzhou Zhu, Xijun Li, Jia Zeng, Jianye HAO, Bin Li, Feng Wu: Rethinking Branching on Exact Combinatorial Optimization Solver: The First Deep Symbolic Discovery Framework. ICLR 2024 [pdf]

  8. Chang Liu, Zhichen Dong, Haobo Ma, Weilin Luo, Xijun Li, Bowen Pang, Jia Zeng, Junchi Yan: L2P-MIP: Learning to Presolve for Mixed Integer Programming. ICLR 2024 [pdf]

  9. Zijie Geng, Xijun Li, Jie Wang, Xiao Li, Yongdong Zhang, Feng Wu: A Deep Instance Generative Framework for MILP Solvers Under Limited Data Availability. NeurIPS 2023 (Spotlight) [page] [pdf] [code]

  10. Yang Li, Xinyan Chen, Wenxuan Guo, Xijun Li, Wanqian Luo, Junhua Huang, Hui-Ling Zhen, Mingxuan Yuan, Junchi Yan: HardSATGEN: Understanding the Difficulty of Hard SAT Formula Generation and A Strong Structure-Hardness-Aware Baseline. KDD 2023 [pdf] [code]

  11. Xijun Li, Qingyu Qu, Fangzhou Zhu, Jia Zeng, Mingxuan Yuan, Jie Wang: Accelerating Linear Programming Solving by Exploiting the Performance Variability via Reinforcement Learning. AAAI 2023 workshop on AI to Accelerate Science and Engineering [pdf] [poster] [slides] [video]

  12. Xijun Li*, Zhihai Wang*, Jie Wang, Yufei Kuang, Mingxuan Yuan, Jia Zeng, Yongdong Zhang, Feng Wu: Learning Cut Selection for Mixed-Integer Linear Programming via Hierarchical Sequence Model. ICLR 2023 [pdf] [slides] [code (MindSpore)] [code (PyTorch)] [media]

  13. Han Lu, Zenan Li, Runzhong Wang, Qibing Ren, Xijun Li, Mingxuan Yuan, Jia Zeng, Xiaokang Yang, Junchi Yan: ROCO: A General Framework for Evaluating Robustness of Combinatorial Optimization Solvers on Graphs. ICLR 2023 [pdf] [code]

  14. Hongduo Liu, Peiyu Liao, Mengchuan Zou, Bowen Pang, Xijun Li, Mingxuan Yuan, Tsung-Yi Ho, Bei Yu: Layout Decomposition via Boolean Satisfiability. DAC 2023 [pdf] [slides]

  15. Yiyuan Yang, Rongshang Li, Qiquan Shi, Xijun Li, Gang Hu, Xing Li, Mingxuan Yaun: SGDP: A Stream-Graph Neural Network Based Data Prefetcher. IJCNN 2023 [pdf] [code]

  16. Wenxuan Guo, Hui-Ling Zhen, Xijun Li#, Mingxuan Yuan, Yaohui Jin, Junchi Yan: Machine Learning Methods in Solving the Boolean Satisfiability Problem. Machine Intelligence Research Journal [pdf]

  17. Jiayi Zhang, Chang Liu, Xijun Li#, Hui-Ling Zhen, Junchi Yan, Mingxuan Yuan: A Survey for Solving Mixed Integer Programming via Machine Learning. Neurocomputing Journal [pdf]

  18. Qingyu Qu, Kexin Liu, Xijun Li, Yunfan Zhou, Jinhu Lv: Satellite Observation and Data-Transmission Scheduling using Imitation Learning based on Mixed Integer Linear Programming. IEEE Transactions on Aerospace and Electronic Systems, TAES [pdf]

  19. Ji Zhang, Xijun Li#, Xiyao Zhou, Mingxuan Yuan, Zhuo Cheng, Keji Huang, Yifan Li: L-QoCo: learning to optimize cache capacity overloading in storage systems. DAC 2022: 379-384 (CORE A, CCF A) [pdf]

  20. Xijun Li*#, Yunfan Zhou*, Jinhong Luo, Mingxuan Yuan, Jia Zeng, Jianguo Yao: Learning to Optimize DAG Scheduling in Heterogeneous Environment. MDM 2022: 137-146 (CORE A, CCF C) [pdf] [video] [slides]

  21. Qingyu Qu, Xijun Li#, Yunfan Zhou; Yordle: An Efficient Imitation Learning for Branch and Bound. NeurIPS 2021 ML4CO Competition [pdf]

  22. Xijun Li, Weilin Luo, Mingxuan Yuan, Jun Wang, Jiawen Lu, Jie Wang, Jinhu Lü, Jia Zeng: Learning to Optimize Industry-Scale Dynamic Pickup and Delivery Problems. ICDE 2021: 2511-2522 (CORE A, CCF A) [pdf] [slides]

  23. Yingtian Tang, Han Lu, Xijun Li#, Lei Chen, Mingxuan Yuan, Jia Zeng: Learning-Aided Heuristics Design for Storage System. SIGMOD 2021: 2597-2601 (CORE A*, CCF A) [pdf] [video]

  24. Zhenkun Wang, Hui-Ling Zhen, Jingda Deng, Qingfu Zhang, Xijun Li, Mingxuan Yuan, Jia Zeng: Multiobjective Optimization-Aided Decision-Making System for Large-Scale Manufacturing Planning. IEEE Transactions on Cybernetics 52(8): 8326-8339 (2022) (CORE A, CCF B) [pdf]

  25. Hui-Ling Zhen, Zhenkun Wang, Xijun Li, Qingfu Zhang, Mingxuan Yuan, Jia Zeng; Accelerate the optimization of large-scale manufacturing planning using game theory. Complex & Intelligent Systems; 1-12 [pdf]

  26. Xijun Li, Mingxuan Yuan, Di Chen, Jianguo Yao, Jia Zeng: A Data-Driven Three-Layer Algorithm for Split Delivery Vehicle Routing Problem with 3D Container Loading Constraint. KDD 2018: 528-536 (CORE A*, CCF A) [pdf] [video] [media]

  27. Xijun Li, Jianguo Yao, Mingxuan Yuan, Jia Zeng: A Two-Layer Algorithmic Framework for Service Provider Configuration and Planning with Optimal Spatial Matching. CIKM 2018: 2273-2281 (CORE A, CCF B) [pdf]

  28. Xijun Li, Jianguo Yao, Xue Liu, Haibing Guan: A First Look at Information Entropy-Based Data Pricing. ICDCS 2017: 2053-2060 (CORE A, CCF B) [pdf]

  29. 李希君, 上海, 上海交通大学: 基于信息熵的数据交易定价研究 [master thesis]

Software

  1. Machine Learning Insides OptVerse AI Solver: Design Principles and Applications [page]

  2. AI助力求解性能持续全面突破,诺亚方舟助力天筹求解器登顶5项权威榜单 [page]

Preprint

  1. Xijun Li, Qingyu Qu, Fangzhou Zhu, Jia Zeng, Mingxuan Yuan, Kun Mao, Jie Wang: Learning to Reformulate for Linear Programming. [arxiv]

  2. Xijun Li*, Yunfan Zhou*, Ji Zhang: PASCAL - A Learning-aided Cooperative Bandwidth Control Policy for Hierarchical Storage Systems. [arxiv]

  3. Jie Wang, Zijie Geng, Xijun Li, Jianye Hao, Yongdong Zhang, Feng Wu: G2MILP: Learning to Generate Mixed-Integer Linear Programming Instances for MILP Solvers [techrxiv]

  4. Yufei Kuang, Xijun Li*, Jie Wang, Fangzhou Zhu, Meng Lu, Zhihai Wang, Jia Zeng, Houqiang Li, Yongdong Zhang, Feng Wu: Accelerate Presolve in Large-Scale Linear Programming via Reinforcement Learning. [arxiv]

  5. Yunfan Zhou*, Xijun Li*#, Qingyu Qu: Offline Reinforcement Learning with Adaptive Behavior Regularization. [arxiv]

  6. Qingyu Qu, Xijun Li*, Yunfan Zhou, Jia Zeng, Mingxuan Yuan, Jie Wang, Jinhu Lv, Kexin Liu, Kun Mao: An Improved Reinforcement Learning Algorithm for Learning to Branch. [arxiv]

  7. Zhihai Wang, Lei Chen, Jie Wang, Xing Li, Yinqi Bai, Xijun Li, Mingxuan Yuan, Jianye Hao, Yongdong Zhang, Feng Wu: A Circuit Domain Generalization Framework for Efficient Logic Synthesis in Chip Design [arxiv]

  8. Haoyang Liu, Yufei Kuang, Jie Wang, Xijun Li, Yongdong Zhang, Feng Wu: Promoting Generalization for Exact Solvers via Adversarial Instance Augmentation. [arxiv]

  9. Wenxuan Guo, Hui-Ling Zhen, Xijun Li, Mingxuan Yuan, Yaohui Jin, Junchi Yan: Machine Learning Methods in Solving the Boolean Satisfiability Problem. [arxiv]

  10. Jiayi Zhang, Chang Liu, Xijun Li, Hui-Ling Zhen, Junchi Yan, Mingxuan Yuan: A Survey for Solving Mixed Integer Programming via Machine Learning. [arxiv]

  11. Jianye Hao, Jiawen Lu, Xijun Li, Xialiang Tong, Xiang Xiang, Mingxuan Yuan, Hankz Hankui Zhuo: Introduction to The Dynamic Pickup and Delivery Problem Benchmark–ICAPS 2021 Competition. [arxiv]

  12. Xinyan Chen, Wenxuan Guo, Yang Li, Wanqian Luo, Hui-Ling Zhen, Xijun Li, Mingxuan Yuan, Junchi Yan: Proceedings of SAT Competition 2022: Solver and Benchmark Descriptions [pdf] [link] where ‘*’ indicates the equal contribution and ‘#’ stands for corresponding author.

Patents

  1. CN116661676A 一种带宽控制方法、数据处理系统及相关设备:李希君,周云帆,李文思,张霁,袁明轩

  2. CN111738409A 一种资源调度的方法及其相关设备:李希君,罗威林,陆佳文,袁明轩

  3. CN116468099A 一种模型结构的优化方法及装置:李希君,朱方舟,甄慧玲,付小津,陆梦,曾嘉,袁明轩

  4. CN115496247A 一种业务数据的处理方法以及装置:李希君,郝晓田,袁明轩,郝建业,曾嘉

  5. CN117370715A 基于云计算技术的目标函数求解方法、装置和计算设备:李希君,李建树,王治海,曾嘉

  6. 92041897CN02 一种数学规划实例生成的方法、系统和电子设备:李希君,安志武,朱方舟,耿子介,王杰

  7. CN117371674A 目标规划问题的求解方法、选择节点的方法及装置:李希君,杨沐明,匡宇飞,曾嘉

  8. CN116933908A 一种计算机任务处理方法及其相关设备:陆梦,甄慧玲,李希君,朱方舟,袁明轩,曾嘉

  9. CN111915060A 组合优化任务的处理方法以及处理装置: 甄慧玲,王振坤,李希君,张青富,袁明轩

  10. CN114237835A 一种任务求解方法及其装置:朱方舟,罗万千,甄慧玲,李希君,袁明轩,曾嘉

  11. CN112818280B 一种信息处理方法以及相关设备:甄慧玲,王振坤,张青富,李希君,韩雄威

  12. CN114117715A 多目标任务优化的方法与装置:王振坤,甄慧玲,李希君,张青富,袁明轩

  13. CN116050522A 预求解配置方法及装置:罗威林,庞博文,李希君,刘畅,严骏驰,曾嘉

  14. CN116483633A 一种数据增广方法及相关装置:黄俊华,罗万千,李希君,甄慧玲,郦洋,严骏驰

Competitions

  1. 3rd Place, Parallel Track, SAT Competition 2023 [result] [solution]

  2. 1st Place, Student Learderboard in Dual Track, NeurIPS 2021 ML4CO Competition [result] [media] [solution]

  3. Meritorious Award, Mathematical Contest in Modeling (MCM) 2014, Consortium for Mathematics and Its Applications [cerification]

Invited Talks

Experience

Education

Selected Awards

Services

Program Committee Member

Journal Reviewer

Interns and Collaborators

Supervised and Co-supervised Students