郑海斌

郑海斌，男，中共党员，博士，现任浙江工业大学助理研究员，主要从事人工智能安全、大数据挖掘、计算机视觉图像处理、对抗攻防方向研究，在ICSE、FSE、ASE、ISSTA、IEEE TDSC、IEEE TIFS、IEEE TCSS、Information Sciences、Computers & Security、计算机研究与发展、软件学报、自动化学报等国内外重要会议和期刊上发表论文20余篇。担任AI Plus、Digital Twins and Applications期刊青年编委。主持国家自然科学基金等项目。

一作论文

1. Haibin Zheng, Gan Xu, Jinyin Chen, Zhenyang Yu. BAIFT: A Behavior-Aware Framework for Intersectional Fairness Testing of LLMs[C]. Proceedings of the 49th International Conference on Software Engineering. Dublin: IEEE/ACM, 2027: 1-12. (CCF-A会议)

2. Haibin Zheng, Jinyin Chen, Haibo Jin. CertPri: Certifiable Prioritization for Deep Neural Networks via Movement Cost in Feature Space[C], IEEE/ACM International Conference on Automated Software Engineering (ASE) 2023. (CCF-A会议)

3. Haibin Zheng, Zhiqing Chen, Tianyu Du, Xuhong Zhang, Yao Cheng, Shouling Ji, Jingyi Wang and Jinyin Chen*. NeuronFair: Interpretable White-Box Fairness Testing through Biased Neuron Identification[C]. In Proceedings of the 44th International Conference on Software Engineering (ICSE '22). Association for Computing Machinery, New York, NY, USA, 1519–1531. (CCF-A会议)

4. Haibin Zheng, Jinyin Chen, Hang Du, Weipeng Zhu, Shouling Ji, and Xuhong Zhang*. GRIP-GAN: An Attack-Free Defense based on General Robust Inverse Perturbation[J]. IEEE Transactions on Dependable and Secure Computing, 2022: 19 (6), 4204-4224. (CCF-A期刊)

5. Haibin Zheng, Jinyin Chen, Tao Liu, Yao Cheng, Zhao Wang, Yun Wang, Lan Gao, Shouling Ji, Xuhong Zhang. DP-poison: Poisoning federated learning under the cover of differential privacy[J]. ACM Transactions on Privacy and Security, 2024, 28(1): 1-28.(CCF-B期刊)

6. Haibin Zheng, Xiaohao Li, Jinyin Chen, Jianfeng Dong, Yan Zhang, Changting Lin*. One4All: Manipulate One Agent to Poison the Cooperative Multi-Agent Reinforcement Learning[J]. Computers & Security, 2023: 124, 1-15. (SCI 2区, CCF-B期刊)

7. Haibin Zheng, Jinyin Chen, Wenchang Shangguan, Zhaoyan Ming, Xing Yang, Zhijun Yang. GONE: A Generic O(1) NoisE Layer for Protecting Privacy of Deep Neural Networks[J]. Computers & Security, 2023. (SCI 2区, CCF-B期刊, IF: 5.105)

8. Haibin Zheng, Haiyang Xiong, Haonan Ma, Guohan Huang, Jinyin Chen*. Link-Backdoor: Backdoor Attack on Link Prediction via Node Injection [J]. IEEE Transactions on Computational Social Systems, 2023. (SCI 2区)

9. Haibin Zheng, Haiyang Xiong, Jinyin Chen*, Haonan Ma, Guohan Huang. Motif-Backdoor: Rethinking the Backdoor Attack on Graph Neural Networks via Motifs [J]. IEEE Transactions on Computational Social Systems, 2023. (SCI 2区)

10. Haibin Zheng, Tao Liu, Rongchang Li and Jinyin Chen*. PoE: Poisoning Enhancement Through Label Smoothing in Federated Learning [J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2023 (SCI 2区)

11. 郑海斌,陈晋音,章燕,张旭鸿,葛春鹏,刘哲,欧阳亦可,纪守领*. 面向自然语言处理的对抗攻防与鲁棒性分析综述[J]. 计算机研究与发展, 2021, 58(8): 1727-1750. （CCF-A中文）

通讯作者论文

1. Jinyin Chen, Haibin Zheng*, Tao Liu, Jiawei Liu, Yao Cheng, Xuhong Zhang, and Shouling Ji. EdgePro: Edge Deep Learning Model Protection via Neuron Authorization[J]. IEEE Transactions on Dependable and Secure Computing (TDSC), DOI: 10.1109/TDSC.2024.3365730, 2024. (CCF-A期刊)

2. Jinyin Chen, Chengyu Jia, Yunjie Yan, Jie Ge, Haibin Zheng*, Yao Cheng. A miss is as good as a mile: Metamorphic testing for deep learning operators[C]. Proceedings of the ACM on Software Engineering, 2024, 1(FSE): 2005–2027.(CCF-A会议)

3. Jinyin Chen, Jie Ge, Haibin Zheng*. ActGraph: Prioritization of Test Cases Based on Deep Neural Network Activation Graph[J], Automated Software Engineering, 2023. (CCF-B会议，SCI)

4. Jinyin Chen, Guohan Huang, Haibin Zheng*, Shanqing Yu, Wenrong Jiang and Chen Cui. Graph-Fraudster: Adversarial Attacks on Graph Neural Network Based Vertical Federated Learning[J]. IEEE Transactions on Computational Social Systems (TCSS), 2022. (SCI)

5. Jinyin Chen, Guohan Huang, Haibin Zheng*, Dunjie Zhang, Xiang Lin. Graphfool: Targeted Label Adversarial Attack on Graph Embedding[J]. IEEE Transactions on Computational Social Systems (TCSS), 2022. (SCI)

6. 李荣昌,刘涛,郑海斌*,陈晋音,刘振广,基于最大最小策略的纵向联邦学习隐私保护方法,自动化学报,2022 (CCF-A中文EI）

博士期间论文（学生一作）

1. Jinyin Chen, Haibin Zheng, Wenchang Shangguan, Liangying Liu, Shouling Ji. ACT-Detector: Adaptive Channel Transformation-based Light-Weighted Detector for Adversarial Attacks[J]. Information Sciences (INS), 2021, 564: 163-192. DOI: 10.1016/j.ins.2021.01.035 (SCI 1区, TOP期刊, IF: 8.233)

2. Jinyin Chen, Haibin Zheng, Hui Xiong, Shijing Shen, Mengmeng Su. MAG-GAN: Massive Attack Generator via GAN[J]. Information Sciences (INS), 2020, 536: 67-90. DOI: 10.1016/j.ins.2020.04.019. (SCI 1区, TOP期刊, IF: 8.233)

3. Jinyin Chen, Haibin Zheng, Hui Xiong, Ruoxi Chen, Tianyu Du, Zhen Hong, Shouling Ji. FineFool: A Novel DNN Object Contour Attack on Image Recognition based on the Attention Perturbation Adversarial Technique [J]. Computers & Security (COSE), 2021, 104: 102220. DOI: 10.1016/j.cose.2021.102220. (SCI 2区, CCF-B, IF: 4.573)

4. Jinyin Chen, Haibin Zheng, Ruoxi Chen, Hui Xiong. RCA-SOC: A Novel Adversarial Defense by Refocusing on Critical Areas and Strengthening Object Contours[J]. Computers & Security (COSE), 2020, 96: 101916. DOI: 10.1016/j.cose.2020.101916. (SCI 2区, CCF-B, IF: 5.105)

5. Jinyin Chen, Haibin Zheng, Hui Xiong, Yangyang Wu, Xiang Lin, Shiyan Ying, Qi Xuan. DGEPN-GCEN2V: A New Framework for Mining GGI and Its Application in Biomarker Detection[J]. SCIENCE CHINA Information Sciences (SCIS), 2019, 61: 050108. DOI: 10.1007/s11432-018-9704-7. (SCI 2区, CCF-B, IF: 2.731)

6. Jinyin Chen, Haibin Zheng, Xiang Lin, Yangyang Wu, Mengmeng Su. A Novel Image Segmentation Method Based on Fast Density Clustering Algorithm[J]. Engineering Applications of Artificial Intelligence (EAAI), 2018, 73: 92-110. DOI: 10.1016/j.engappai.2018.04.023. (SCI 2区, CCF-C, IF: 3.640)

7. Jinyin Chen, Haibin Zheng, Mengmeng Su, Tianyu Du, Changting Lin, Shouling Ji. Invisible Poisoning: Highly Stealthy Targeted Poisoning Attack[C]. Proceedings of the 15th International Conference on Information Security and Cryptology (Inscrypt 2019). Springer, Cham, 2019: 173-198. (EI: 20202208719133) (最佳论文)

授权专利

1. 郑海斌 等. 一种自蒸馏模型后门移除的任务执行方法、电子设备、介质，ZL 202411732164.3

2. 郑海斌 等. 一种基于多级算子变异的强化学习框架安全性检测方法，ZL 202211358344.0

3. 郑海斌 等. 基于智能化渗透测试的战略蜜罐部署防御方法，ZL 202211508855.6

4. 郑海斌 等. 一种基于情感倾向的可解释推荐方法，ZL 202211357975.0

5. 郑海斌 等. 基于循环神经网络的红外目标检测模型频闪隐身方法，ZL 202211366759.2

6. 陈晋音, 郑海斌, 等. 基于灰度峰值的裂缝位置查找方法. ZL 201610115242.4
   

7. 陈晋音, 郑海斌, 等. 一种基于快速密度聚类算法的图像分割方法. ZL 201610887803.2

8. 陈晋音, 郑海斌, 等. 基于深度学习和进化计算的特征基因选择方法. ZL 201711336535.6

9. 陈晋音, 郑海斌, 等. 基于可视图算法的基因表达时序数据分类方法. ZL 201810386434.8

10. 陈晋音, 郑海斌, 等. 基于流形学习与闭环深度卷积双网络模型的基因特征提取方法. ZL 201710355356.0

11. 陈晋音, 郑海斌, 等. 基于对抗式生成网络的深度学习对抗性攻击防御方法. ZL 201810141210.0.

12. 陈晋音, 郑海斌, 等. 基于数据流聚类的癌症亚型精准发现与演化分析方法. ZL 201710355340.X

13. 陈晋音, 郑海斌, 等. 基于RNN的基因调控网络构建与动态差异性分析方法. ZL 201710355357.5

14. 陈晋音, 郑海斌, 等. 一种基于LSTM的分级式无线信号调制类型识别方法. ZL 201910089087.7

15. 陈晋音, 郑海斌, 等. 一种基于特征重映射的对抗样本防御方法和应用. ZL 202010115357.X

16. 陈晋音, 成凯回, 郑海斌. 基于通道检测的面向无线信号识别的对抗攻击防御方法与装置. ZL 201911031922.8

1. 国家自然科学基金青年项目，智能算法的偏见测试与纠偏研究（62406286），2025.01-2027.12，主持。

2. 北京生命科技研究院有限公司（重点实验室）一般课题，基于深度学习的细胞比对算法研究与平台构建（2024200CD0210），2024.09-2026.08，主持。

3. 数据安全防护与智能治理教育部重点实验室重点课题，面向大模型的公平性测试和偏见消除方法研究（SCUSAKFKT202402Z），2024.10-2026.09，主持。

4. 浙江大学，智能感知算法模型基础环境构建和数据加工测试，2023.11-2024.12，主持。

5. 浙江大学，后门算法模型的基础环境构建，2024.06-2024.12，主持。

6. 杭州熔合智造科技有限公司，智能算法模型产权防护技术研发，2023.06-2025.05，主持。

1 第八届“西湖科创杯”中国研究生智慧城市技术与创意设计大赛一等奖

项目名称：基于深度学习的智能驾驶安全研究

时间：2023年03月

本科生课程：人工智能概论、人工智能导论、人工智能原理及应用、深度学习

留学生课程：Software Engineering

2026.12-2028.12，AI Plus青年编，委参与论文审稿、专题策划、优质稿源组织与推荐及学术宣传交流

2025.12-2027.12，Digital Twins and Applications青年编委，参与论文审稿、专题策划、优质稿源组织与推荐及学术宣传交流

2024.10.09-10.11，2024智能视觉产业发展大会专题秘书，会议组织与学术服务

2026.07-2026.11，AAAI 2027程序委员会委员，承担投稿论文的同行评审与评审讨论，参与论文遴选和质量把关