Abstract
Quantum Secure Multiparty Computation (QMPC) is a new way to use cryptography that lets a lot of people work together to compute a function over their private inputs while keeping those inputs secret, even if attackers have quantum computing power. As AI becomes more common in fields like healthcare, banking, and self-driving cars, secure collaborative learning becomes more and more important. But the rise of quantum computers is a big threat to the encryption methods that are commonly used in modern multiparty computation protocols. QMPC is a very secure and scalable way to protect privacy in AI. It does this by combining quantum-safe oblivious transfer (QOT), quantum key distribution (QKD), and quantum oblivious linear evaluation (qOLE).
This study shows a full QMPC architecture for AI systems that are both federated and decentralized. It lets you train and use private neural networks and is safe from quantum attacks because it uses safe aggregation and encrypted processing protocols. The architecture lets AI do a lot of things, like private set intersection, federated learning, and collaborative model training, without giving away user data. A thorough security analysis based on post-quantum assumptions and the universal composability paradigm makes sure that the system can withstand attackers who are adaptive, colluding, or quantum-capable. When tested against datasets like MNIST and CIFAR-10, the protocol gets results that are as accurate as plaintext models with very little extra work on the computer or network. Experiments show that when the accuracy of the model drops by less than 1%, all data privacy is still protected.
This study looks at how QMPC might help make AI safer in sensitive areas like medical diagnostics and financial forecasts, as well as how it might help companies follow data protection laws like GDPR and HIPAA. Auditability and consent procedures are two of the legal and moral issues that need to be thought about. The study also talks about problems with scalability and suggests an ideal communication plan and methods like circuit pruning, quantization, and hybrid classical-quantum communication models. We look at what QMPC could do in relation to secure federated analytics, real-time collaborative robotics, and processing encrypted smart contracts.
In conclusion, QMPC for privacy-preserving AI is no longer just a theory; it is now a necessary, possible, and realistic way of doing things in a world that is quickly moving toward quantum supremacy. By using quantum-resistant encryption and distributed machine learning together, QMPC creates a way to get to safe, understandable, and decentralized AI that will work in the future. This study sets the stage for future progress at the crossroads of post-quantum security and collaborative intelligence.
PDF
Reference
[1] Shor, P. W. (1994). Factorization and discrete logarithms are two methods used in quantum computing. Proceedings of the 35th Annual Symposium on Computer Science Foundations.
[2] O. Goldreich (2004). Volume 2 of Foundations of Cryptography. Press from Cambridge University.
[3] Gentry, C. (2009). cryptography with perfect lattices that works for everything. STOC.
[4] E. Kashefi, J. Fitzsimons, and A. Broadbent (2009). Quantum blindness in universal computing. IEEE FOCS.
[5] A., Ben-Or, M., and Goldwasser, S. (1988). Theorems of completeness for distributed computation that can handle faults and doesn't use cryptography. STOC.
[6] Damgård, I., Nielsen, J. B., and Cramer, R. (2015). Safe secret sharing and multiparty computation. Cambridge University Press.
[7] Liang, M., Wu, X., and Qin, Z. (2023). A survey of federated learning that is quantum-secure. ACM Computing Surveys.
[8] Alagic, G., and others (2020). This is an update on the first step of the NIST process to standardize post-quantum cryptography. NIST.
[9] L. Grover (1996). A quick way to search through quantum mechanical databases. STOC.
[10] Rivest, R. L., Shamir, A., and Adleman, L. (1978). A way to get digital signatures and public-key cryptography systems. ACM letters.
[11] O. Regev (2005). Cryptography, random linear codes, and learning with errors in lattices. The ACM Journal.
[12] Halevi, S. and Shoup, V. (2014). Algorithms for HElib. cryptocurrency.
[13] Z and V. Vaikuntanathan. Brakerski (2011). efficient fully homomorphic encryption from (standard) LWE. FOCS.
[14] Albrecht, M. R., and others (2015). About the real difficulty of learning from your mistakes. The Journal of Mathematical Cryptology.
[15] Baum, C., and others (2020). MPC in the brain: new ideas that can be used with ZK and non-interactive protocols. ACM CCS.
[16] Morimae, T. and Fujii, K. (2013). A regular person using quantum blind computing. Physical Assessment A.
[17] Damgård, I. and Nielsen, J. B. (2007). Multiparty computation that can be scaled up and is completely safe. cryptocurrency.
[18] Wei, R., Xie, S., and Yu, Y. (2022). A federated learning system that is safe after quantum computers come out and uses multi-party homomorphic encryption. sciences related to information.
[19] J. Kilian (1988). The idea of oblivious transfer cryptography was born. STOC.
[20] Childs and friends, A. M. (2013). quantum walks that speed up algorithms by a lot. SIAM J. Comput.
[21] Qin, Z., and others (2021). A study of federated learning that is private and safe. IEEE Communications Surveys and Tutorials.
[22] Zuo, C., Liu, Y., and Chen, L. (2021). Machine learning that protects privacy along with quantum security. Computer systems of the future.
[23] Z. Brakerski. (2012). Encryption that is fully homomorphic from regular GapSVP without changing the modulus. cryptocurrency.
[24] M. Fitzi, N. Gisin, and U. Maurer (2001). The quantum answer to the Byzantine agreement problem. letters of physical review.
[25] Z. Jafargholi and D. Wichs (2016). Publicly verifiable effective zero-knowledge based on (optimal) lattice assumptions. TCC.
[26] Yu and others (2019). toward a safe IoT that uses AI and blockchain. IEEE Network.
[27] Boneh, D. and Lipton, R. J. (1995). Quantum cryptanalysis of hidden linear functions. cryptocurrency.
[28] N. Gisin and coworkers, 2002. Quantum encryption. Rev. Mod. in a physical way.
[29] Rass, S. and Schauer, S. (2020). There are a number of problems with quantum computing right now as we work toward safe AI. Springer.
[30] Abspoel, D., and others (2020). ring-LWE's secure and fast MPC. Europe.
[31] Fehr, S., and Schaffner, C. (2010). writing quantum code in a classical setting. Conference on the Theory of Cryptography.
[32] Dwork, C. and Roth, A. (2014). the basis of algorithms' different levels of privacy. Theoretical Computer Science: Its Foundations and Trends.
[33] Yang and others (2023). Federated learning with blockchain and quantum-safe guarantees. IEEE Transactions on Engineering and Network Science.
[34] Wang, T. and Chen, H. (2020). Homomorphic encryption makes it possible to safely use the cloud and share data. Cloud Computing Journal.
[35] Al-Bassam, M., et al. (2018). Chainspace is a smart contract platform that is split into shards. The NDSS Conference.
[36] Bennink, R. S., and others (2002). A 10 Gb/s clock rate for quantum key distribution. letters to the editor of Physical Review.
[37] Franklin, M. and Boneh, D. (2001). Weil pairing is a way to encrypt data based on identity. cryptocurrency.
[38] O. and Micciancio. Regev. (2009). Cryptography based on lattices. cryptography based on quantum theoryKashefi, E. and Fitzsimons, J. F. (2017). Blind quantum computing that can be verified without any conditions. Physical Assessment A.
[39] Yao, A. C. (1982). rules for safe calculations. FOCS.
[40] Zhang and friends (2020). machine learning that protects privacy with homomorphic encryption. sciences that deal with information.
[41] Sun and friends (2021). Using lattice encryption to make federated learning safe after quantum computing. IEEE Transactions on Security and Information Forensics.
[42] Zhang and others (2019). PQ-MPC is short for Post-Quantum Secure Multiparty Computation. arXiv:1904.05349.
[43] Bost and others, 2015. using machine learning to sort encrypted data. NDSS.
[44] Chen, L. K., and others (2022). AI that is safe from quantum attacks and can work with others. what machine intelligence is like.
[45] S. Wang and others (2020). Federated learning that can be used in edge computing systems with limited resources. IEEE Journal on Specific Topics in Communications.
[46] Zhao, R., and others (2023). Quantum-enhanced secure federated learning. IEEE Transactions on Quantum Engineering.
[47] Y. Liang and others (2020). AI with secure multiparty computation for smart grids. IEEE Transactions on Smart Grids.
[48] Wang and his coworkers (2023). A look at quantum MPC and the problems it faces. Computer systems of the future.
[49] Green, M., Hohenberger, S., and Waters, B. (2011). We are hiring someone else to decode ABE ciphertext. USENIX's safety.
[50] Please tell me if you would like these references to be in your paper in a certain style, like IEEE, MLA, APA, or another.