{"type":"ScholarlyArticle","title":"Privacy-Preserving UCB Decision Process Verification via zk-SNARKs","authors":[{"name":"Xikun Jiang","slug":"xikun-jiang","given":"Xikun","family":"Jiang","affiliation":"Department of Computer Science , Shanghai Jiao Tong University , China; Department of Computer Science , University of Copenhagen , Denmark"},{"name":"He Lyu","slug":"he-lyu","given":"He","family":"Lyu","affiliation":"Department of Computer Science , University of Copenhagen , Denmark"},{"name":"Chenhao Ying","slug":"chenhao-ying","given":"Chenhao","family":"Ying","affiliation":"Department of Computer Science , Shanghai Jiao Tong University , China; Shanghai Jiao Tong University (Wuxi) Blockchain Advanced Research Center , China"},{"name":"Yibin Xu","slug":"yibin-xu","given":"Yibin","family":"Xu","affiliation":"Department of Computer Science , University of Copenhagen , Denmark"},{"name":"Boris Düdder","slug":"boris-dudder","given":"Boris","family":"Düdder","affiliation":"Department of Computer Science , University of Copenhagen , Denmark"},{"name":"Yuan Luo","slug":"yuan-luo","given":"Yuan","family":"Luo","affiliation":"Department of Computer Science , Shanghai Jiao Tong University , China; Shanghai Jiao Tong University (Wuxi) Blockchain Advanced Research Center , China"}],"abstract":"With the increasingly widespread application of machine learning, how to strike a balance between protecting the privacy of data and algorithm parameters and ensuring the verifiability of machine learning has always been a challenge. This study explores the intersection of reinforcement learning and data privacy, specifically addressing the Multi-Armed Bandit (MAB) problem with the Upper Confidence Bound (UCB) algorithm. We introduce zkUCB, an innovative algorithm that employs the Zero-Knowledge Succinct Non-Interactive Argument of Knowledge (zk-SNARKs) to enhance UCB. zkUCB is carefully designed to safeguard the confidentiality of training data and algorithmic parameters, ensuring transparent UCB decision-making.\r\n\r\nExperiments highlight zkUCB's superior performance, attributing its enhanced reward to judicious quantization bit usage that reduces information entropy in the decision-making process. zkUCB's proof size and verification time scale linearly with the execution steps of zkUCB. This showcases zkUCB's adept balance between data security and operational efficiency. This approach contributes significantly to the ongoing discourse on reinforcing data privacy in complex decision-making processes, offering a promising solution for privacy-sensitive applications.","doi":"10.24963/ijcai.2024/652","url":"https://ijcai.org/proceedings/2024/652","pdf":"https://ijcai.org/proceedings/2024/0652.pdf","markdown":"https://ijcai.org/proceedings/2024/652.md","bibtex":"https://ijcai.org/proceedings/2024/bibtex/652","volume":{"slug":"2024","title":"Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2024","url":"https://ijcai.org/proceedings/2024/","year":2024,"location":"Jeju, Korea","dates":"3-9 August 2024","editors":"Kate Larson","isbn":"978-1-956792-04-1","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2024"},"section":"Multidisciplinary Topics and Applications","track":"Main Track","number":652,"pages":{"first":5900,"last":5908},"keywords":["Machine Learning: ML: Trustworthy machine learning","Machine Learning: ML: Multi-armed bandits","Multidisciplinary Topics and Applications: MTA: Security and privacy"],"videos":[],"references":[{"position":1,"raw":"[Afsar et al., 2022] M Mehdi Afsar, Trafford Crump, and Behrouz Far. Reinforcement learning based recommender systems: A survey. ACM Computing Surveys, 55(7):1-38, 2022.","title":"Reinforcement learning based recommender systems: A survey","authors":"Trafford Mehdi Afsar, Behrouz Crump, Far","year":"2022","venue":"ACM Computing Surveys","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Ameko et al., 2020] Mawulolo K Ameko, Miranda L Beltzer, Lihua Cai, Mehdi Boukhechba, Bethany A Teach- man, and Laura E Barnes. Offline contextual multi-armed bandits for mobile health interventions: A case study on emotion regulation. In Proceedings of the 14th ACM Con- ference on Recommender Systems, pages 249-258, 2020.","title":"Offline contextual multi-armed bandits for mobile health interventions: A case study on emotion regulation","authors":"Miranda L Mawulolo K Ameko, Lihua Beltzer, Mehdi Cai, Bethany A Boukhechba, Laura E Teachman, Barnes","year":"2020","venue":"Proceedings of the 14th ACM Conference on Recommender Systems","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Aziz et al., 2021] Maryam Aziz, Emilie Kaufmann, and Marie-Karelle Riviere. On multi-armed bandit designs for dose-finding clinical trials. The Journal of Machine Learn- ing Research, 22(1):686-723, 2021.","title":"On multi-armed bandit designs for dose-finding clinical trials","authors":"Maryam Aziz, Emilie Kaufmann, Marie-Karelle Riviere","year":"2021","venue":"The Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Azize and Basu, 2022] Achraf Azize and Debabrota Basu. When privacy meets partial information: A refined anal- ysis of differentially private bandits. Advances in Neural Information Processing Systems, 35:32199-32210, 2022.","title":"Achraf Azize and Debabrota Basu. When privacy meets partial information: A refined analysis of differentially private bandits","authors":"Basu Azize","year":"2022","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Baelde et al., 2021] David Baelde, Stéphanie Delaune, Charlie Jacomme, Adrien Koutsos, and Solène Moreau. An interactive prover for protocol verification in the com- putational model. In 2021 IEEE Symposium on Security and Privacy (SP), pages 537-554. IEEE, 2021.","title":"An interactive prover for protocol verification in the computational model","authors":"David Baelde, Stéphanie Delaune, Charlie Jacomme, Adrien Koutsos, Solène Moreau","year":"2021","venue":"IEEE Symposium on Security and Privacy (SP)","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Chen et al., 2022] Thomas Chen, Hui Lu, Teeramet Kunpit- taya, and Alan Luo. A review of zk-snarks. arXiv preprint arXiv:2202.06877, 2022.","title":"A review of zk-snarks","authors":"Thomas Chen, Hui Lu, Teeramet Kunpittaya, Alan Luo","year":"2022","venue":null,"doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Gao et al., 2022] Guoju Gao, Sijie Huang, He Huang, Mingjun Xiao, Jie Wu, Yu-E Sun, and Sheng Zhang. Combination of auction theory and multi-armed bandits: Model, algorithm, and application. IEEE Transactions on Mobile Computing, 2022.","title":"Combination of auction theory and multi-armed bandits: Model, algorithm, and application","authors":"Gao","year":"2022","venue":"IEEE Transactions on Mobile Computing","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Ghodsi et al., 2017] Zahra Ghodsi, Tianyu Gu, and Sid- dharth Garg. Safetynets: Verifiable execution of deep neu- ral networks on an untrusted cloud. Advances in Neural Information Processing Systems, 30, 2017.","title":"Safetynets: Verifiable execution of deep neural networks on an untrusted cloud","authors":"Zahra Ghodsi, Tianyu Gu, Siddharth Garg","year":"2017","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Groth, 2016] Jens Groth. On the size of pairing-based non-interactive arguments. In Advances in Cryptology- EUROCRYPT 2016: 35th Annual International Confer- ence on the Theory and Applications of Cryptographic Techniques, Vienna, Austria, May 8-12, 2016, Proceed- ings, Part II 35, pages 305-326. Springer, 2016.","title":"On the size of pairing-based non-interactive arguments","authors":"Jens Groth, Groth","year":"2016","venue":"Advances in Cryptology-EUROCRYPT 2016: 35th Annual International Conference on the Theory and Applications of Cryptographic Techniques","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Jawurek et al., 2013] Marek Jawurek, Florian Kerschbaum, and Claudio Orlandi. Zero-knowledge using garbled cir- cuits: how to prove non-algebraic statements efficiently. In Proceedings of the 2013 ACM SIGSAC conference on Computer & communications security, pages 955-966, 2013.","title":"Zero-knowledge using garbled circuits: how to prove non-algebraic statements efficiently","authors":"Marek Jawurek, Florian Kerschbaum, Claudio Orlandi","year":"2013","venue":"Proceedings of the 2013 ACM SIGSAC conference on Computer & communications security","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Kim et al., 2023] Taechan Kim, Hyesun Kwak, Dongwon Lee, Jinyeong Seo, and Yongsoo Song. Asymptotically faster multi-key homomorphic encryption from homomor- phic gadget decomposition. In Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communica- tions Security, pages 726-740, 2023.","title":"Asymptotically faster multi-key homomorphic encryption from homomorphic gadget decomposition","authors":"Kim","year":"2023","venue":"Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communications Security","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Kiran et al., 2021] B Ravi Kiran, Ibrahim Sobh, Victor Tal- paert, Patrick Mannion, Ahmad A Al Sallab, Senthil Yoga- mani, and Patrick Pérez. Deep reinforcement learning for autonomous driving: A survey. IEEE Transactions on In- telligent Transportation Systems, 23(6):4909-4926, 2021.","title":"Deep reinforcement learning for autonomous driving: A survey","authors":"Ibrahim Ravi Kiran, Victor Sobh, Patrick Talpaert, Ahmad A Al Mannion, Senthil Sallab, Patrick Yogamani, Pérez","year":"2021","venue":"IEEE Transactions on Intelligent Transportation Systems","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Lee et al., 2024] Seunghwa Lee, Hankyung Ko, Jihye Kim, and Hyunok Oh. vcnn: Verifiable convolutional neural network based on zk-snarks. IEEE Transactions on De- pendable and Secure Computing, 2024.","title":"vcnn: Verifiable convolutional neural network based on zk-snarks","authors":"Lee","year":"2024","venue":"IEEE Transactions on Dependable and Secure Computing","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Li et al., 2020] Jing Li, Xiaohui Kuang, Shujie Lin, Xu Ma, and Yi Tang. Privacy preservation for machine learning training and classification based on homomorphic encryp- tion schemes. Information Sciences, 526:166-179, 2020.","title":"Privacy preservation for machine learning training and classification based on homomorphic encryption schemes","authors":"Jing Li, Xiaohui Kuang, Shujie Lin, Xu Ma, Yi Tang","year":"2020","venue":"Information Sciences","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Liu et al., 2021] Tianyi Liu, Xiang Xie, and Yupeng Zhang. Zkcnn: Zero knowledge proofs for convolutional neural network predictions and accuracy. In Proceedings of the 2021 ACM SIGSAC Conference on Computer and Com- munications Security, pages 2968-2985, 2021.","title":"Zkcnn: Zero knowledge proofs for convolutional neural network predictions and accuracy","authors":"Tianyi Liu, Xiang Xie, Yupeng Zhang","year":"2021","venue":"Proceedings of the 2021 ACM SIGSAC Conference on Computer and Communications Security","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Mouris and Tsoutsos, 2021] Dimitris Mouris and Nektar- ios Georgios Tsoutsos. Zilch: A framework for deploying transparent zero-knowledge proofs. IEEE Transactions on Information Forensics and Security, 16:3269-3284, 2021.","title":"Dimitris Mouris and Nektarios Georgios Tsoutsos. Zilch: A framework for deploying transparent zero-knowledge proofs","authors":"Tsoutsos Mouris","year":"2021","venue":"IEEE Transactions on Information Forensics and Security","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Patil et al., 2021] Vishakha Patil, Ganesh Ghalme, Vineet Nair, and Yadati Narahari. Achieving fairness in the stochastic multi-armed bandit problem. The Journal of Machine Learning Research, 22(1):7885-7915, 2021.","title":"Achieving fairness in the stochastic multi-armed bandit problem","authors":"Vishakha Patil, Ganesh Ghalme, Vineet Nair, Yadati Narahari","year":"2021","venue":"The Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Sun et al., 2020] Yi Sun, Qian Liu, Xingyuan Chen, and Xuehui Du. An adaptive authenticated data structure with privacy-preserving for big data stream in cloud. IEEE Transactions on Information Forensics and Security, 15:3295-3310, 2020.","title":"An adaptive authenticated data structure with privacy-preserving for big data stream in cloud","authors":"Yi Sun, Qian Liu, Xingyuan Chen, Xuehui Du","year":"2020","venue":"IEEE Transactions on Information Forensics and Security","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Wahby et al., 2018] Riad S Wahby, Ioanna Tzialla, Abhi Shelat, Justin Thaler, and Michael Walfish. Doubly- efficient zksnarks without trusted setup. In 2018 IEEE Symposium on Security and Privacy (SP), pages 926-943. IEEE, 2018.","title":"Doublyefficient zksnarks without trusted setup","authors":"Ioanna Riad S Wahby, Abhi Tzialla, Justin Shelat, Michael Thaler, Walfish","year":"2018","venue":"IEEE Symposium on Security and Privacy (SP)","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Weng et al., 2021] Chenkai Weng, Kang Yang, Xiang Xie, Jonathan Katz, and Xiao Wang. Mystique: Efficient con- versions for {Zero-Knowledge} proofs with applications to machine learning. In 30th USENIX Security Symposium (USENIX Security 21), pages 501-518, 2021.","title":"Mystique: Efficient conversions for {Zero-Knowledge} proofs with applications to machine learning","authors":"Chenkai Weng, Kang Yang, Xiang Xie, Jonathan Katz, Xiao Wang","year":"2021","venue":"30th USENIX Security Symposium (USENIX Security 21)","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Xing et al., 2023] Zhibo Xing, Zijian Zhang, Jiamou Liu, Ziang Zhang, Meng Li, Liehuang Zhu, and Giovanni Rus- sello. Zero-knowledge proof meets machine learning in verifiability: A survey. arXiv preprint arXiv:2310.14848, 2023.","title":"Zero-knowledge proof meets machine learning in verifiability: A survey","authors":"Zhibo Xing, Zijian Zhang, Jiamou Liu, Ziang Zhang, Meng Li, Liehuang Zhu, Giovanni Russello","year":"2023","venue":null,"doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Yang and Wang, 2022] Kang Yang and Xiao Wang. Non- interactive zero-knowledge proofs to multiple verifiers. In International Conference on the Theory and Application of Cryptology and Information Security, pages 517-546. Springer, 2022.","title":"Noninteractive zero-knowledge proofs to multiple verifiers","authors":"Kang Yang, Xiao Wang","year":"2022","venue":"International Conference on the Theory and Application of Cryptology and Information Security","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Yu et al., 2021] Chao Yu, Jiming Liu, Shamim Nemati, and Guosheng Yin. Reinforcement learning in healthcare: A survey. ACM Computing Surveys (CSUR), 55(1):1-36, 2021.","title":"Reinforcement learning in healthcare: A survey","authors":"Chao Yu, Jiming Liu, Shamim Nemati, Guosheng Yin","year":"2021","venue":"ACM Computing Surveys (CSUR)","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Zhang et al., 2020] Jiaheng Zhang, Zhiyong Fang, Yupeng Zhang, and Dawn Song. Zero knowledge proofs for deci- sion tree predictions and accuracy. In Proceedings of the 2020 ACM SIGSAC Conference on Computer and Com- munications Security, pages 2039-2053, 2020.","title":"Zero knowledge proofs for decision tree predictions and accuracy","authors":"Zhang","year":"2020","venue":"Proceedings of the 2020 ACM SIGSAC Conference on Computer and Communications Security","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Zhao et al., 2020] Hui Zhao, Mingjun Xiao, Jie Wu, Yun Xu, He Huang, and Sheng Zhang. Differentially private unknown worker recruitment for mobile crowdsensing us- ing multi-armed bandits. IEEE Transactions on Mobile Computing, 20(9):2779-2794, 2020.","title":"Differentially private unknown worker recruitment for mobile crowdsensing using multi-armed bandits","authors":"Hui Zhao, Mingjun Xiao, Jie Wu, Yun Xu, He Huang, Sheng Zhang","year":"2020","venue":"IEEE Transactions on Mobile Computing","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Zhao et al., 2021] Lingchen Zhao, Qian Wang, Cong Wang, Qi Li, Chao Shen, and Bo Feng. Veriml: Enabling integrity assurances and fair payments for machine learning as a service. IEEE Transactions on Parallel and Distributed Systems, 32(10):2524-2540, 2021.","title":"Veriml: Enabling integrity assurances and fair payments for machine learning as a service","authors":"Lingchen Zhao, Qian Wang, Cong Wang, Qi Li, Chao Shen, Bo Feng","year":"2021","venue":"IEEE Transactions on Parallel and Distributed Systems","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}