Jiasi Weng (翁嘉思)

Research Interests

Trustworthiness-enhancing techniques for collaborative machine learning, such as verifiable computation, privacy protection, incentive mechanisms as well as blockchain.

About Me

Jiasi Weng currently is an associate professor in Jinan University. Jiasi Weng received her PhD degree in College of Computer Science from Jinan University in 2021, and her advisor is Prof. Jian Weng. From Dec. 2021 to Aug. 2022, she was a Postdoc supervised by Prof. Cong Wang in City University of Hong Kong, and from Oct. 2019 to Oct. 2020, she worked as a Research Assistant. Jiasi defended her doctoral thesis titled "Study on the Key Technologies of Blockchain-Based Secure and Trustworthy Collaborative Machine Learning".

Publications

• Tao Huang, Jian Weng, Jiasi Weng, Min-Rong Chen, Ming Li. SecFloatPlus: More Accurate Floating-Point Meets Secure Two-Party Computation. To be accepted by ProvSec 2024.

• Jiasi Weng, Shenglong Yao, Yuefeng Du, Junjie Huang, Jian Weng*, Cong Wang. Proof of Unlearning: Definitions and Instantiation. https://arxiv.org/abs/2210.11334.pdf, accepted by IEEE Transactions on Information Forensics and Security (IEEE TIFS), 2024.

• Jiasi Weng, Jian Weng*, Gui Tang, Anjia Yang, Ming Li, Jia-Nan Liu. pvCNN: Privacy-Preserving and Verifiable Convolutional Neural Network Testing. IEEE Transactions on Information Forensics and Security (IEEE TIFS), 18: 2218-2233, 2023.

• Jiasi Weng, Jian Weng*, Jilian Zhang, Ming Li, Yue Zhang, Weiqi Luo. DeepChain: Auditable and Privacy-Preserving Deep Learning with Blockchain-based Incentive. IEEE Transactions on Dependable and Secure Computing (IEEE TDSC), 18(5):2438 - 2455, 2021. 

Essential Science Indicators (ESI) "Highly Cited Paper" and "Hot Paper"

• Jiasi Weng, Jian Weng*, Chengjun Cai, Hongwei Huang, Cong Wang. Golden Grain: Building a Secure and Decentralized Model Marketplace for MLaaS. IEEE Transactions on Dependable and Secure Computing (IEEE TDSC), 19(5):3149-3167, 2022.

• Jiasi Weng, Jian Weng*, Hongwei Huang, Chengjun Cai, Cong Wang. FedServing: A Federated Prediction Serving Framework Based on Incentive Mechanism. The 40th International Conference on Computer Communications (INFOCOM 2021), May 10-13, 2021

• Jiasi Weng, Jian Weng*, Yue Zhang, Weiqi Luo, Weiming Lan. BENBI: Scalable and Dynamic Access Control on the Northbound Interface of SDN-based VANET. IEEE Transactions on Vehicular Technology (IEEE TVT), 68(1):822-831, 2019.

• Jin-Hua Chen, Min-Rong Chen, Guo-Qiang Zeng, Jia-Si Weng. BDFL: a Byzantine-Fault-Tolerance Decentralized Federated Learning Method for Autonomous Vehicle. IEEE Transactions on Vehicular Technology (IEEE TVT), 70(9): 8639-8652, 2021.

• Xin Chen, Jian Weng, Wei Lu, Jiaming Xu, Jiasi Weng. Deep Manifold Learning Combined with Convolutional Neural Networks for Action Recognition. IEEE transactions on neural networks and learning systems (IEEE TNNLS), 29(9): 3938-3952, 2017.

Patents

• 翁健, 翁嘉思 (2/5), 刘家男, 魏凯敏, 罗伟其. 基于区块链的软件定义网络控制层安全机制构建方法. 20190917, ZL201710389296.4, 中国发明专利. 已授权，已成果转化.

• 翁健, 张悦, 魏林锋, 翁嘉思 (2/5).一种Android 应用程序代码保护机制鉴别方法. 20200117, ZL201710047861.9, 中国发明专利, 已授权.

• 翁健, 李勇标, 李明, 李宇娴, 翁嘉思 (2/5). 吴永东. 一种基于区块链及代理重加密的数据传输方法及系统. 20220906, ZL202011145558.0, 中国发明专利, 已授权.

• Jian Weng, Ming Li, Jiasi Weng (2/5), Yue Zhang, Anjia Yang, Weiqi Luo, Secure Blockchain-based Vehicular Digital Forensics，20220111, No. US 11223487, 美国专利, 已授权.

• Jian Weng, Jiasi Weng (2/5), Ming Li, Yue Zhang, Jilian Zhang, Weiqi Luo, Auditable privacy protection deep learning platform construction method based on block chain incentive mechanism，20200618, No. US 16702786, 美国专利, 已授权.

• Jian Weng, Yue zhang, Jiasi Weng (2/5), Ming Li, Weiqi Luo, Method and system for maintaining privacy and traceability of blockchain-based system，20220214, No. US 16824063, 美国专利, 已授权.

• Jian Weng, Zhijian Shao, Yue Zhang, Ming Li, Jiasi Weng (2/5), Yongdong Wu, Weiqi Luo, Peripheral-free secure pairing protocol by randomly switching power, 20220301, No. US 11265722, 美国专利, 已授权.

Projects

• 项目名称：面向多方协作机器学习的安全与隐私保护技术研究，项目来源和类别：国家自然科学基金委员会青年科学基金项目，项目编号：62302192，执行时间：2024.01.01-2026.12.31，经费：30万元，获批时间：2023.12.04，项目角色：主持

• 项目名称：面向联邦学习的数据删除安全与隐私保护研究，项目来源和类别：广东省自然科学基金委员会面上项目，项目编号：2024A1515010086，执行时间：2024.01.01-2026.12.31，经费：15万元，获批时间：2023.08.24，项目角色：主持

• 项目名称：基于密码学的多方隐私计算关键技术研究，项目来源和类别：国家自然科学基金委员会重点项目，项目编号：62332007，执行时间：2024.01.01-2028.12.31，经费：231万元，获批时间：2023.08.24，项目角色：第一主要参与人

• 项目名称：安全可信的协作机器学习关键技术研究，项目来源和类别：中央高校基本科研业务费（自然科学）项目，执行时间：2023.06.01-2024.12.31，经费：9万，项目角色：主持

• 项目名称： 数据安全与隐私保护下的机器学习技术, ，项目来源和类别：中华人民共和国科学技术部,国家重点研发计划，项目编号：2021ZD0112802，执行时间：2021.12.01-2025.11.01，经费：100万元，项目角色：项目骨干

Research Story