曹晓锋

职称:副教授、博士生导师

毕业院校:悉尼科技大学

email:xiaofeng.cao.uts@gmail.com

个人主页:https://xiaofengcaoml.github.io

研究方向:主动学习理论、非欧几何建模、泛化分析

个人简介

曹晓锋,男,吉林大学人工智能学院副教授,长期从事人工智能和机器学习理论的基础研究工作,曾在澳大利亚人工智能研究院|悉尼科技大学(澳洲top 1, 国际top 10 AI Center)取得博士学位并担任研究助理职位,累计发表超10篇学术论文,含多篇CCF A类顶级国际学术会议和IEEE trans系列中科院一区top期刊。主要研究方向为机器学习理论的基础问题,具体包括主动学习理论、非欧几何建模、泛化分析及相关应用等方面,长期担任人工智能和机器学习领域旗舰期刊和会议的审稿人,如Journal of Machine Learning Research (JMLR)、Machine Learning Journal (MLJ)、Journal of Artificial Intelligence Research (JAIR)、Artificial Intelligence Journal (AIJ)、IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI)、ICML、NeurIPS等。

从2020年至今,尝试解决机器学习理论大师Corinna Cortes教授(谷歌研究院副总裁,SVM发明人之一,谷歌索引7万+)在主动学习理论中提出的“关于黑盒学习者如何使用误差分歧修剪假设空间”这一难题,合作导师为国际知名统计机器学者Ivor W. Tsang教授(澳大利亚人工智能研究院研究主任,新加坡前沿人工智能研究中心主任,谷歌索引2万+),IEEE Fellow。

工作经历

2020.03 —— 至      今    吉林大学人工智能学院  副教授

2021.01 —— 至      今    悉尼科技大学  研究助理

教育经历

2017.09 —— 2021.01 澳大利亚人工智能研究院|悉尼科技大学计算机科学  博士

2014.09 —— 2017.06 郑州大学计算机科学与技术 硕士

2010.09 —— 2014.06 郑州大学计算机科学与技术 本科

科研项目

2023.01-2025.12 面向黑盒的机器教学收敛研究,国家自然基金项目,在研,负责人。

2023.01-2025.12 吉林大学励新青年教师计划,在研,负责人。

2024.01-2024.12 飞秒激光永久光存储数据写入过程的智能化研究 ,省科技厅面上项目 在研,主要参与。

论文选

你可能对以下一个或多个主题感兴趣,部分主题可能涉及机器学习理论、黑盒优化等较为数学化的内容:

§ Hyperbolic   Geometry (双曲空间和黎曼几何的建模与优化)

[1] Xiaofeng Cao, Ivor W. Tsang.  Distribution Disagreement   via Lorentzian Focal Representation, IEEE Transactions on Pattern Analysis and   Machine Intelligence. (T-PAMI为模式分析旗舰期刊,IF=17.861)

[2] Xiaofeng Cao, Ivor W. Tsang. Learning Hyperbolic Fréchet Mean   for Deep Representations. Review in IEEE TNNLS. 神经学习汇刊.

[3] Xiaofeng Xu, Ivor W. Tsang, Xiaofeng Cao, et al. Learning   image-specific attributes by hyperbolic neighborhood graph propagation. IJCAI   2019.

[4] Yang Tao and Xiaofeng Cao. Perturbation   Elimination via Homeomorphic Manifold Tubes, in progress.

§ Machine   Teaching/ Black-box/Bayesian Optimization (机器教学/黑盒/贝叶斯优化)

[1] Xiaofeng   Cao, Ivor W. Tsang. Distribution-based Machine Teaching for a Black-box,   Artificial Intelligence, to be accepted.

[2] Xiaofeng Cao, Ivor W. Tsang. On the Geometry of Deep Bayesian   Active Learning, T-PAMI, under review.

[3] Chen Zhang, Xiaofeng Cao. Pseudo-Iterative Machine Teaching. Pseudo-Iterative   Machine Teaching, in progress.

[4] Xiaofeng Cao# and Yaming Guo#. Black-box Teaching an Active   Learner, in progress.

§ Active   Learning Theory and Its Generalization Analysis (主动学习理论与其泛化分析)

[1] Xiaofeng Cao, Ivor W. Tsang. Shattering distribution for active   learning, TNNLS, 2020.

[2] Xiaofeng Cao, Ivor W. Tsang, Jianliang Xu. Cold-start Active   Sampling via $\gamma$-Tube, T-CYB, 2021.

[3] Xiaofeng   Cao. A structured perspective of volumes on active learning. Neurocomputing,   377: 200-212, 2020. (唯一作者)

[4] Xiaofeng Cao. A divide-and-conquer approach to geometric   sampling for active learning, Expert Systems with Applications, 140, 2020. (唯一作者)

§ Learning on Small Data (新主题:Small data is the future of AI)

[1] Xiaofeng Cao, Ivor W. Tsang.  Learning on Small data via Minimizing   Hyperspherical Energy. Review in T-PAMI.

[2] Learning on Small Data: Transfer the Future of Artificial   Intelligence to Now, Survey work, in progress.

§ Applications of Data Mining and Image Analysis (涵盖部分硕士研究生工作)

[1] Xiaofeng Cao et al. Multidimensional Balance-Based Cluster   Boundary Detection for High-Dimensional Data, TNNLS, 30(6): 1867-1880, 2019.

[2] Xiaofeng Cao et al. BorderShift: toward optimal MeanShift   vector for cluster boundary detection in high-dimensional data, Pattern   Analysis and Applications 22(3): 1015-1027, 2019.

[3] Xiaofeng Cao. High-dimensional cluster boundary detection using   directed Markov tree, Pattern Analysis and Applications, 2020. (唯一作者)

[4] Baozhi Qiu, Xiaofeng Cao*. Clustering boundary detection for   high dimensional space based on space inversion and Hopkins statistics.   Knowledge-Based Systems 98 (2016): 216-225. (通讯作者,硕士研究生工作)

[5] Zenglin Shi, Le Zhang, Yun Liu, Xiaofeng Cao et al. Crowd   counting with deep negative correlation learning. CVPR 2018.

[6] Xiangli Li, Xiaofeng Cao*, Baozhi   Qiu. Clustering boundary pattern discovery for high dimensional space base on   matrix model. Acta Automatica Sinica, 43(11), pp.1962-1972, 2017. (通讯作者,硕士研究生工作)

注:如果你有意与我一同工作,并且拟攻读硕士研究生,你可能会直接参与以上课题。如果你有意攻读博士,我们将一起探索更为基础和前沿的机器学习内容,可能包括Meta-Learning、Distribution   Optimization, 等。你可能与我、Ivor   Tsang一起工作。Ivor Tsang是国际知名机器学者,他是NeurIPS 2021 Exp Chair, ACML 2021 Co-Chair, ICML 2021 Senior   Area Chair,同时也是JMLR、MLJ、T-PAMI、JAIR等机器学习和人工智能旗舰期刊的Editor/Associate Editor。

社会兼职

Conference Reviewer/Program Committee: ACML 2021, ICML 2021,   NeurIPS 2021.

Journal Reviewer: T-PAMI, MLJ, TNNLS.

返回上一层