目前尚有研究生招生名额,也欢迎本科生实习,欢迎邮件联系!
研究方向
机器学习, 随机优化, 大模型优化.
工作经历
-
2025.10 至今, 南京理工大学.
- 教授, 计算机科学与工程学院.
教育经历
-
2020.09 - 2025.06, 博士, 计算机科学与技术, 南京大学.
- 2023.10 - 2024.05, 联合培养博士研究生, 新加坡国立大学.
-
2016.09 - 2020.06, 学士, 计算机科学与技术(实验班), 西安交通大学.
- 2019.01 - 2019.06, 交换生, 美国加州大学伯克利分校. [证书][成绩单]
第一作者论文 (所有论文列表)
- ICML 2026 Convergence Analysis of the Lion Optimizer in Centralized and Distributed Settings
In Proceedings of the 43rd International Conference on Machine Learning (ICML 2026), to appear, 2026. - TPAMI 2026 Optimizing Unnormalized Statistical Models through Compositional Optimization [PDF]
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 48(2): 1949 - 1960, 2026. - TPAMI 2025 Revisiting Stochastic Multi-Level Compositional Optimization [PDF]
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 47(7): 5613 - 5624, 2025. - Journal of Software 2025 Normalized Adaptive Variance Reduction Method [PDF]
Journal of Software, 36(11): 4893 - 4905, 2025. - NeurIPS 2024 Adaptive Variance Reduction for Stochastic Optimization under Weaker Assumptions [PDF]
In Advances in Neural Information Processing Systems 37 (NeurIPS 2024), pages 22047 - 22080, 2024. - NeurIPS 2024 Efficient Sign-Based Optimization: Accelerating Convergence via Variance Reduction [PDF]
In Advances in Neural Information Processing Systems 37 (NeurIPS 2024), pages 33891 - 33932, 2024. - ICML 2024 Projection-Free Variance Reduction Methods for Stochastic Constrained Multi-Level Compositional Optimization [PDF]
In Proceedings of the 41st International Conference on Machine Learning (ICML 2024), pages 21962 - 21987, 2024. - ICML 2023 Learning Unnormalized Statistical Models via Compositional Optimization [PDF]
In Proceedings of the 40th International Conference on Machine Learning (ICML 2023), pages 15105 - 15124, 2023. - NeurIPS 2022 Multi-block-Single-probe Variance Reduced Estimator for Coupled Compositional Optimization [PDF, Supplementary]
In Advances in Neural Information Processing Systems 35 (NeurIPS 2022), pages 32499 - 32511, 2022. - ICML 2022 Optimal Algorithms for Stochastic Multi-Level Compositional Optimization [PDF]
In Proceedings of the 39th International Conference on Machine Learning (ICML 2022), pages 10195 - 10216, 2022.
奖励与荣誉
- 南京大学优秀毕业生, 2025
- 博士研究生国家奖学金, 2024
- 南京大学优秀研究生标兵, 2024
- NeurIPS优秀审稿人, 2024
- 博士研究生国家奖学金, 2023
- 南京大学优秀研究生标兵, 2023
- LAMDA英才奖, 2023
- 腾讯奖学金, 2022
- 南京大学优秀研究生, 2022
- 兴业银行奖学金, 2021
- 南京大学优秀研究生, 2021
- DeeCamp人工智能训练营总冠军 (奖金10万元)
- 西安交通大学优秀毕业生, 2020
项目
- 面向大模型的高性能随机优化方法研究. 国家自然科学基金青年科学基金项目(C类), 2027-2029.
- 资源受限场景下的大模型优化算法研究. 江苏省自然科学基金(青年基金项目), 2027-2029.
- 复合损失函数的分布式学习. 江苏省研究生科研创新计划, 2024-2025.
学术服务
- 会议领域主席: NeurIPS, ICLR.
- 会议审稿人: ICML, NeurIPS, ICLR, AAAI, AISTATS.
- 期刊审稿人:
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- IEEE Transactions on Information Forensics and Security
- IEEE Transactions on Evolutionary Computation
- Scientific Reports
- Machine Learning
- Applied Numerical Mathematics
- Information Sciences
- Neurocomputing
- Transactions on Machine Learning Research
- Journal of Machine Learning Research
- Journal of Scientific Computing