当前位置: X-MOL首页全球导师 国内导师 › 刘金艳

个人简介

刘金艳,准聘教授,北京理工大学高层次青年人才。于香港大学计算机系获得博士学位(导师: Dr. Zhiyi Huang, Prof. Siu Ming Yiu),博士期间曾在Nanyang Technological University, Singapore (Host: Dr. Xiaohui Bei)和Georgia Institute of Technology, USA (Host: Dr. Rachel Cummings)访学。 研究基础为理论计算机(Theoretical Computer Science, TCS)中的隐私保护和算法博弈论,具体方向包括差分隐私,联邦学习,拍卖机制和资源分配等。在理论的基础上,也进行应用性研究:包括人工智能学习算法的隐私保护,信息资源的价值衡量与交易,互联网大数据杀熟的应对策略,和社会资源的公平分配等。在计算机科学国际顶级会议/期刊NeurIPS, AAAI, AI, PODS等发表多篇学术论文,主持一项国家自然科学基金青年项目,担任NeurIPS, ICML, TKDE, ICLR, ESA, ITCS等计算机会议/期刊审稿人。获得人工智能国际顶会AAAI 2020唯一最佳学生论文奖。 承担科研情况 2022.01-2024.12,国家自然科学基金青年项目,62102026,基于隐私保护的拍卖机制和在线学习算法的研究,30万元,主持 2019.11-2023.12,北京理工大学高层次人才科研启动计划(海外高层次青年人才),100万元,主持 所获奖励 人工智能国际顶会AAAI 2020最佳学生论文奖 社会兼职 NeurIPS, ICML,ICLR, ESA, ISAAC, ITCS, FAW, TPDP等计算机会议审稿人 TKDE期刊审稿人 IJTCS-FAW程序委员会委员

研究领域

隐私保护,联邦学习,机制设计,资源分配等

近期论文

查看导师最新文章 (温馨提示:请注意重名现象,建议点开原文通过作者单位确认)

Truthful Generalized Linear Models Yuan Qiu, Jinyan Liu and Di Wang The 18th Conference on Web and Internet Economics (WINE), 2022. Fair Division of Mixed Divisible and Indivisible Goods Xiaohui Bei, Zihao Li, Jinyan Liu, Shengxin Liu and Xinhang Lu Artificial Intelligence (AI, CCF A类), 2021. Fair Division of Mixed Divisible and Indivisible Goods Xiaohui Bei, Zihao Li, Jinyan Liu, Shengxin Liu and Xinhang Lu Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI, CCF A类), 2020. (最佳学生论文奖). Privacy-preserving Crowd-guided AI Decision-making in Ethical Dilemmas Teng Wang, Jun Zhao, Han Yu, Jinyan Liu, Xinyu Yang, Xuebing Ren, and Shuyu Shi Twenty-Eighth ACM International Conference on Information and Knowledge Management (CIKM, CCF B类), 2019. Learning Optimal Reserve Price against Non-myopic Bidders Zhiyi Huang, Jinyan Liu and Xiangning Wang Advances in Neural Information Processing Systems (NeurIPS, CCF A类), 2018. Optimal Differentially Private Algorithms for k-Means Clustering Zhiyi Huang and Jinyan Liu Proceeding of the ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Databases Systems (PODS, CCF B类,理论数据库国际顶会), 2018.

推荐链接
down
wechat
bug