Privacy-Preserving UCB Decision Process Verification via zk-SNARKs (2404.12186v3)
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. Experiments 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.
- Reinforcement learning based recommender systems: A survey. ACM Computing Surveys, 55(7):1–38, 2022.
- Offline contextual multi-armed bandits for mobile health interventions: A case study on emotion regulation. In Proceedings of the 14th ACM Conference on Recommender Systems, pages 249–258, 2020.
- On multi-armed bandit designs for dose-finding clinical trials. The Journal of Machine Learning Research, 22(1):686–723, 2021.
- When privacy meets partial information: A refined analysis of differentially private bandits. Advances in Neural Information Processing Systems, 35:32199–32210, 2022.
- An interactive prover for protocol verification in the computational model. In 2021 IEEE Symposium on Security and Privacy (SP), pages 537–554. IEEE, 2021.
- A review of zk-snarks. arXiv preprint arXiv:2202.06877, 2022.
- Combination of auction theory and multi-armed bandits: Model, algorithm, and application. IEEE Transactions on Mobile Computing, 2022.
- Safetynets: Verifiable execution of deep neural networks on an untrusted cloud. Advances in Neural Information Processing Systems, 30, 2017.
- Jens Groth. On the size of pairing-based non-interactive arguments. In Advances in Cryptology–EUROCRYPT 2016: 35th Annual International Conference on the Theory and Applications of Cryptographic Techniques, Vienna, Austria, May 8-12, 2016, Proceedings, Part II 35, pages 305–326. Springer, 2016.
- Zero-knowledge using garbled circuits: how to prove non-algebraic statements efficiently. In Proceedings of the 2013 ACM SIGSAC conference on Computer & communications security, pages 955–966, 2013.
- Asymptotically faster multi-key homomorphic encryption from homomorphic gadget decomposition. In Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communications Security, pages 726–740, 2023.
- Deep reinforcement learning for autonomous driving: A survey. IEEE Transactions on Intelligent Transportation Systems, 23(6):4909–4926, 2021.
- vcnn: Verifiable convolutional neural network based on zk-snarks. IEEE Transactions on Dependable and Secure Computing, 2024.
- Privacy preservation for machine learning training and classification based on homomorphic encryption schemes. Information Sciences, 526:166–179, 2020.
- Zkcnn: Zero knowledge proofs for convolutional neural network predictions and accuracy. In Proceedings of the 2021 ACM SIGSAC Conference on Computer and Communications Security, pages 2968–2985, 2021.
- Zilch: A framework for deploying transparent zero-knowledge proofs. IEEE Transactions on Information Forensics and Security, 16:3269–3284, 2021.
- Achieving fairness in the stochastic multi-armed bandit problem. The Journal of Machine Learning Research, 22(1):7885–7915, 2021.
- 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.
- Doubly-efficient zksnarks without trusted setup. In 2018 IEEE Symposium on Security and Privacy (SP), pages 926–943. IEEE, 2018.
- Mystique: Efficient conversions for {{\{{Zero-Knowledge}}\}} proofs with applications to machine learning. In 30th USENIX Security Symposium (USENIX Security 21), pages 501–518, 2021.
- Zero-knowledge proof meets machine learning in verifiability: A survey. arXiv preprint arXiv:2310.14848, 2023.
- 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.
- Reinforcement learning in healthcare: A survey. ACM Computing Surveys (CSUR), 55(1):1–36, 2021.
- Zero knowledge proofs for decision tree predictions and accuracy. In Proceedings of the 2020 ACM SIGSAC Conference on Computer and Communications Security, pages 2039–2053, 2020.
- Differentially private unknown worker recruitment for mobile crowdsensing using multi-armed bandits. IEEE Transactions on Mobile Computing, 20(9):2779–2794, 2020.
- 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.