Paper 2025/507

Scalable Zero-knowledge Proofs for Non-linear Functions in Machine Learning

Meng Hao, University of Electronic Science and Technology of China
Hanxiao Chen, University of Electronic Science and Technology of China
Hongwei Li, University of Electronic Science and Technology of China
Chenkai Weng, Northwestern University
Yuan Zhang, University of Electronic Science and Technology of China
Haomiao Yang, University of Electronic Science and Technology of China
Tianwei Zhang, Nanyang Technological University
Abstract

Zero-knowledge (ZK) proofs have been recently explored for the integrity of machine learning (ML) inference. However, these protocols suffer from high computational overhead, with the primary bottleneck stemming from the evaluation of non-linear functions. In this paper, we propose the first systematic ZK proof framework for non-linear mathematical functions in ML using the perspective of table lookup. The key challenge is that table lookup cannot be directly applied to non-linear functions in ML since it would suffer from inefficiencies due to the intolerably large table. Therefore, we carefully design several important building blocks, including digital decomposition, comparison, and truncation, such that they can effectively utilize table lookup with a quite small table size while ensuring the soundness of proofs. Based on these blocks, we implement complex mathematical operations and further construct ZK proofs for current mainstream non-linear functions in ML such as ReLU, sigmoid, and normalization. The extensive experimental evaluation shows that our framework achieves 50 ∼ 179× runtime improvement compared to the state-of-the-art work, while maintaining a similar level of communication efficiency.

Metadata
Available format(s)
PDF
Category
Cryptographic protocols
Publication info
Published elsewhere. Major revision. USENIX Security 2024
Keywords
Zero-knowledge proofsZKML
Contact author(s)
menghao303 @ gmail com
History
2025-03-20: approved
2025-03-18: received
See all versions
Short URL
https://ia.cr/2025/507
License
Creative Commons Attribution
CC BY

BibTeX

@misc{cryptoeprint:2025/507,
      author = {Meng Hao and Hanxiao Chen and Hongwei Li and Chenkai Weng and Yuan Zhang and Haomiao Yang and Tianwei Zhang},
      title = {Scalable Zero-knowledge Proofs for Non-linear Functions in Machine Learning},
      howpublished = {Cryptology {ePrint} Archive, Paper 2025/507},
      year = {2025},
      url = {https://eprint.iacr.org/2025/507}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.