Professor He Sun
About Me
He Sun is an Associate Director of State Key Laboratory of Biomedical Imaging Science and System in China, and Director of Center for Algorithms and Learning Theory, Shenzhen Institutes of Advanced Technology (SIAT), Chinese Academy of Sciences (CAS). He received his PhD from Fudan University in 2010, and worked at the Max Planck Institute for Informatics, UC Berkeley, University of Bristol, and University of Edinburgh. His research interests include theoretical computer science, and machine learning theory. He has written over 60 papers and 1 book, and supervised 6 PhD students.
He received the President's Medal of Fudan University (2004), Shanghai Outstanding PhD Thesis Award (2010), Simons-Berkeley Research Fellowship (2014), and has so far received research grants of more than 5 million GBP from different research foundations and industrial partners. In 2020, he was awarded an EPSRC Fellowship of 1.5 million GBP for developing advanced spectral algorithms and Spectral Toolkit of Algorithms for Graphs (STAG), which is an C++ based open-source library for spectral graph algorithms; the Fellowship runs from January 2020 to May 2025.
简介
孙贺， 1984年生，博士，研究员，国家级领军人才，英国工程与物理学基金会会士(EPSRC Fellow)。孙贺于2002年进入复旦大学学习、并用三年时间完成了大学本科四年的学习任务； 由于其优异的学术表现，孙贺在2004年获得复旦大学在校师生和校友的最高荣誉——“复旦大学校长奖”。孙贺在2005年秋起在复旦大学继续深造，并于2009年获得博士学位； 他的博士论文获得上海市优秀博士论文奖。
孙贺在2010年起在德国马普信息科学研究所从事博士后研究，并在2013年8月晋升为高级研究员。2014年，他关于大规模图快速算法的研究获得德国科学基金会（DFG）的经费资助； 同年，孙贺获得Simons-Berkeley Fellow，应邀在美国加州大学伯克利分校从事为期四个月的学术研究。自2015年起，孙贺先后在英国布里斯托大学和爱丁堡大学从事研究与教学工作； 2023年秋，他作为美国加州大学伯克利分校的访问教授，在Simons-Berkeley计算理论研究所开展研究工作。 孙贺在2018年被授予Turing Fellow， 在2020年被英国工程与物理学基金会授予EPSRC Fellow。在过去近20年的时间里， 孙贺先后解决了计算几何、算法谱图论、和人工智能等领域国际公认的学术难题，并作为独立PI获得了累计超过4000万人民币的科研经费资助。此外，他多次担任欧盟科研基金会（ERC）、以色列科学基金会(ISF)、 英国研究与创新基金会（UKRI）等基金委员会的项目评审专家，以及包括ICML（机器学习国际大会）和STOC（计算理论国际大会）在内的计算机科学国际顶级会议的领域主席和程序委员会委员。
孙贺目前担任医学成像科学与技术系统全国重点实验室副主任，并在中国科学院深圳先进技术研究院主持建立算法与学习理论研究中心。
Recent Professional Activities
PC Member of STOC 2026; PC Member of ESA 2026; Area Chair of ICML (2024, 2025, 2026); Senior PC member of AAAI (2025, 2026); Keynote Speaker for COCOA 2024, PDCAT 2024, TAMC 2025
Current and Past PhD Students
Shihong Song, Suranjan De, Ben Jourdan, Steinar Laenen, Peter Macgregor, Bogdan Manghiuc, Luca Zanetti
Publication
Signed Laplacians for constrained graph clustering
J. Stewart Fabila Carrasco, and H. Sun. ICML'25, Spotlight
paperDynamic similarity graph construction with kernel density estimation
S. Laenen, P. Macgregor, and H. Sun. ICML'25
paperCan we measure the impact of a database?
P. Buneman, D. Dosso, M. Lissandrini, G. Silvello, and H. Sun. Communications of the ACM, May 2025
paperDynamic spectral clustering with provable approximation guarantee
S. Laenen, and H. Sun. ICML'24
paperFast approximation of similarity graphs with kernel density estimation
P. Macgregor, and H. Sun. NeurIPS'23, Spotlight
paperIs the algorithmic Kadison-Singer problem hard?
B. Jourdan, P. Macgregor, and H. Sun. ISAAC'23
paperNearly-optimal hierarchical clustering for well-clustered graphs
B. Manghiuc, S. Laenen, and H. Sun. ICML'23
paperThe support of open versus closed random walks
T. Sauerwald, H. Sun, and D. Vagnozzi. ICALP'23
paperFully-dynamic graph sparsifiers against an adaptive adversary.
A. Bernstein, J. van den Brand, M. Gutenberg, D. Nanongkai, T. Saranurak, A. Sidford, and H. Sun. ICALP'22
paperHierarchical clustering: O(1)-approximation for well-clustered graphs
B. Manghiuc, and H. Sun. NeurIPS'21
paperLocal algorithms for finding densely connected clusters
P. Macgregor, and H. Sun. ICML'21, Long Talk.
paperHermitian matrices for clustering directed graphs: insights and applications
M. Cucuringu, H. Li, H. Sun, and L. Zanetti. AISTATS'20
paperHermitian Laplacians and a Cheeger inequality for the Max-2-Lin problem
H. Li, H. Sun, and L. Zanetti. ESA'19
paperDistributed graph clustering and sparsification
H. Sun, and L. Zanetti. ACM Transactions on Parallel Computing, 6(3): 17:1-17:23,2019
paperHuman motion parsing by hierarchical dynamic clustering
Y. Zhang, S. Tang, H. Sun, and H. Neumann. BMVC'18
paperCommunication-optimal distributed clustering
J. Chen, D. Woodruff, H. Sun, and Q. Zhang. NeurIPS'16
paperPartitioning well-clustered graphs: spectral clustering works!
R. Peng, H. Sun, and L. Zanetti. SIAM Journal on Computing'17, COLT'15
paperConstructing linear-sized spectral sparsification in almost-linear time
Y. Lee, and H. Sun. SIAM Journal on Computing'18, FOCS'15
paperBalls into bins via local search: cover time and maximum load
K.Bringmann, T.Sauerwald, A.Stauffer, and H. Sun. Random Structures & Algorithms'16, STACS'14
paperLow randomness rumor spreading via hashing
G. Giakkoupis, T. Sauerwald, H. Sun and P. Woelfel. STACS'12
paperTight bounds for randomized load balancing on arbitrary network topologies
T. Sauerwald, and H. Sun. FOCS'12
paperCounting arbitrary subgraphs in data streams
D.Kane, K.Mehlhorn, T.Sauerwald, and H. Sun. ICALP'12
paperApproximate counting of cycles in streams
M.Manjunath, K.Mehlhorn, K.Panagiotou, and H. Sun. ESA'11
paperMinimum Manhattan network is NP-complete
F.Chin, Z.Guo, and H. Sun. Discrete & Computational Geometry'11, SoCG'09
paper