师资资源
葛冬冬
- 系 别:数据与商务智能系
- 办公电话: 52301361
- 职 称:教授
- 电子邮箱:ddge@sjtu.edu.cn
教师简介
-
上海交通大学智能计算研究院院长、数据与商务智能系特聘教授。主持国家自然科学基金的原创探索,杰出青年,重大基金等项目。担任了国内首个开源数学规划软件LEAVES mathematical programming solver开发,领导了国内首个专业数学优化软件Cardinal Optimizer(COPT)开发,首个专业数学规划和数学优化求解器COPT的开发负责人。在管理与运筹,优化理论,计算机理论科学,机器学习等多个顶级期刊和会议上,如Operations Research, Mathematics of Operation Research, Mathematical Programming, FOCS, SODA, ICML ,NeurIPS等发表过论文。担任过多个国际国内著名期刊的编辑和特约审稿。担任杉数科技首席科学家,深度参与了多个业界合作的重要项目,如与京东,顺丰,滴滴、华为、国网/南网、中石油/中石化/国家管网、上海地铁、黄骅港、上汽通用、南航、波音、谷歌等企业与行业的合作。
研究兴趣:1. 超大规模数学优化问题的理论、算法与软件研发,及其在供应链、制造、交通、能源、量子计算等领域的应用;2. LP,MILP,SDP,SOCP等问题的算法设计、理论分析与软件开发;3. 基于GPU的新一代数学规划算法设计;4. 产GPU的高精度高性能计算数学库函数建设;5. 大模型训练推理中的算法优化,及决策大模型的训练与应用。
数学优化软件开发:1.Cardinal Optimizer(COPT): 领导了国内首个专业数学优化软件开发,目前有线性规划、整数规划、半定规划、二阶锥规划、二阶凸规划、混合整数二阶锥规划、混合整数半正定规划、非线性规划等模块,截止2024年5月,均在美国的第三方测试榜单上排名第一,具体信息请参见:http://plato.asu.edu/bench.html, https://shanshu.ai/solver, 以及 Cardinal Optimizer (COPT) User Guide 2022, https://arxiv.org/abs/2208.14314 2.LEAVES mathematical programming solver: 国内首个开源数学规划软件,2017年发布,包括了线性规划、几何规划等模块。
教育经历:
2004-08 至 2009-08, 斯坦福大学, 管理科学与工程, 博士
1999-08 至 2004-07, 纽约州立大学石溪分校, 数学,计算机,硕士
1995-07 至 1999-07, 南开大学, 数学, 学士
工作经历:
2024至今, 上海交通大学, 安泰经济与管理学院, 特聘教授;智能计算研究院,院长
2013-2023,上海财经大学, 信息管理与工程学院, 教授;交叉科学研究院,院长
2009-2013, 上海交通大学, 安泰经济与管理学院, 讲师,副教授
科学研究
-
科研论文
近三年论文:
1. Early Birds versus Last-Minute Arrivals: Empirical Evidence and Theoretical Analysis of Arrival Time Queueing Game,X Zhao, Y Ding, D Ge, X Xie,Available at SSRN 4955803,2024
2. Solving Integrated Process Planning and Scheduling Problem via Graph Neural Network Based Deep Reinforcement Learning,H Li, H Zhang, Z He, Y Jia, B Jiang, X Huang, D Ge,arXiv preprint arXiv:2409.00968,2024
3. Accelerating Low-Rank Factorization-Based Semidefinite Programming Algorithms on GPU,Q Han, Z Lin, H Liu, C Chen, Q Deng, D Ge, Y Ye,arXiv preprint arXiv:2407.15049,2024
4. An enhanced alternating direction method of multipliers-based interior point method for linear and conic optimization,Q Deng, Q Feng, W Gao, D Ge, B Jiang, Y Jiang, J Liu, T Liu, C Xue, Y Ye, C zhang,INFORMS Journal on Computing,2024
5. ORLM: Training Large Language Models for Optimization Modeling,Z Tang, C Huang, X Zheng, S Hu, Z Wang, D Ge, B Wang,arXiv preprint arXiv:2405.17743,2024
6. Restarted Primal-Dual Hybrid Conjugate Gradient Method for Large-Scale Quadratic Programming,Y Huang, W Zhang, H Li, W Xue, D Ge, H Liu, Y Ye,arXiv preprint arXiv:2405.16160,2024
7. Sketched Newton Value Iteration for Large-Scale Markov Decision Processes,J Liu, C Xie, Q Deng, D Ge, Y Ye,Proceedings of the AAAI Conference on Artificial Intelligence 38 (12), 2024
8. Trust Region Methods For Nonconvex Stochastic Optimization Beyond Lipschitz Smoothness,C Xie, C Li, C Zhang, Q Deng, D Ge, Y Ye,The 38th Annual AAAI Conference on Artificial Intelligence AAAI 2024,2024
9. Learning to Pivot as a Smart Expert,T Liu, S Pu, D Ge, Y Ye,The 38th Annual AAAI Conference on Artificial Intelligence AAAI 2024,2024
10. A Low-Rank ADMM Splitting Approach for Semidefinite Programming,Q Han, C Li, Z Lin, C Chen, Q Deng, D Ge, H Liu, Y Ye,arXiv preprint arXiv:2403.09133,2024
11. Decoupling Learning and Decision-Making: Breaking the Barrier in Online Resource Allocation with First-Order Methods,W Gao, C Sun, C Xue, D Ge, Y Ye,arXiv preprint arXiv:2402.07108,2024
12. Nonlinear modeling and interior point algorithm for the material flow optimization in petroleum refinery,F Dong, D Ge, L Yang, Z Wei, S Guo, H Xu,Electronic Research Archive 32 (2), 915-927,2024
13. A Homogenization Approach for Gradient-Dominated Stochastic Optimization,J Tan, C Xue, C Zhang, Q Deng, D Ge, Y Ye,The Conference on Uncertainty in Artificial Intelligence UAI 2024,2024
14. cuPDLP-C: A Strengthened Implementation of cuPDLP for Linear Programming by C language,H Lu, J Yang, H Hu, Q Huangfu, J Liu, T Liu, Y Ye, C Zhang, D Ge,arXiv preprint arXiv:2312.14832,2023
15. A Universal Trust-Region Method for Convex and Nonconvex Optimization,Y Jiang, C He, C Zhang, D Ge, B Jiang, Y Ye,arXiv preprint arXiv:2311.11489,2023
16. Solving Linear Programs with Fast Online Learning Algorithms,W Gao, D Ge, C Sun, Y Ye
17. ICML'23: Proceedings of the 40th International Conference on Machine Learning,2023
18. Homogeneous Second-Order Descent Framework: A Fast Alternative to Newton-Type Methods,C He, Y Jiang, C Zhang, D Ge, B Jiang, Y Ye,arXiv preprint arXiv:2306.17516,2023
19. Pre-trained Mixed Integer Optimization through Multi-variable Cardinality Branching,Y Chen, W Gao, D Ge, Y Ye,arXiv preprint arXiv:2305.12352,2023
20. Stochastic Dimension-reduced Second-order Methods for Policy Optimization,J Liu, C Xie, Q Deng, D Ge, Y Ye,arXiv preprint arXiv:2301.12174,2023
21. A Homogeneous Second-Order Descent Method for Nonconvex Optimization,C Zhang, D Ge, C He, B Jiang, Y Jiang, C Xue, Y Ye,arXiv preprint arXiv:2211.08212,2022
22. SOLNP+: A Derivative-Free Solver for Constrained Nonlinear Optimization,D Ge, T Liu, J Liu, J Tan, Y Ye,arXiv preprint arXiv:2210.07160,2022
23. Cardinal Optimizer (COPT) user guide,D Ge, Q Huangfu, Z Wang, J Wu, Y Ye,arXiv preprint arXiv:2208.14314,2022
24. Bayesian dynamic learning and pricing with strategic customers,X Chen, J Gao, D Ge, Z Wang,Production and Operations Management 31 (8), 3125-3142,2022
25. DRSOM: A Dimension Reduced Second-Order Method,C Zhang, D Ge, C He, B Jiang, Y Jiang, Y Ye,arXiv preprint arXiv:2208.00208,2022
26. Randomized Branching Strategy in Solving SCUC Model,R Cao, Y Chen, W Gao, J Gao, Y Zhang, C Lu, D Ge,2022 4th International Conference on Power and Energy Technology,2022
27. Hdsdp: Software for semidefinite programming,W Gao, D Ge, Y Ye,arXiv preprint arXiv:2207.13862,2022
28. Optimization and operations research in mitigation of a pandemic,CH Chen, YH Du, DD Ge, L Lei, YY Ye,Journal of the Operations Research Society of China 10 (2), 289-304,2022
29. JD. com: Operations research algorithms drive intelligent warehouse robots to work,H Qin, J Xiao, D Ge, L Xin, J Gao, S He, H Hu, JG Carlsson,INFORMS Journal on Applied Analytics 52 (1), 42-55,2022
30. Fast Online Algorithms for Linear Programming,W Gao, D Ge, C Sun, Y Ye,arXiv preprint arXiv:2107.03570,2021
31. Uncertainty quantification for demand prediction in contextual dynamic pricing,Y Wang, X Chen, X Chang, D Ge,Production and Operations Management 30 (6), 1703-1717,2021
32. From an interior point to a corner point: smart crossover,D Ge, C Wang, Z Xiong, Y Ye,arXiv preprint arXiv:2102.09420,2021
33. A Gradient Descent Method for Estimating the Markov Chain Choice Model,L Fu, DD Ge,Journal of the Operations Research Society of China, 1-11,2021
科研项目
主持国家自然科学基金的原创探索,杰出青年,重大基金等项目。
数学优化软件开发
1.Cardinal Optimizer(COPT): 领导了国内首个专业数学优化软件开发,目前有线性规划、整数规划、半定规划、二阶锥规划、二阶凸规划、混合整数二阶锥规划、混合整数半正定规划、非线性规划等模块,截止2024年5月,均在美国的第三方测试榜单上排名第一,具体信息请参见:http://plato.asu.edu/bench.html, https://shanshu.ai/solver, 以及 Cardinal Optimizer (COPT) User Guide 2022, https://arxiv.org/abs/2208.14314
2.LEAVES mathematical programming solver: 国内首个开源数学规划软件,2017年发布,包括了线性规划、几何规划等模块。
主讲课程
-
智能决策
人工智能与人文社科
高等运筹与优化理论,博士必修
优化理论与物流管理,硕士选修
量化管理科学,新生研讨课
线性与非线性优化,试点班必修课
计算复杂度理论与算法设计