Model training integrity verification method based on trusted execution environment

By constructing an operator runtime prediction model and a trusted execution environment, and combining cryptographic hash functions and quantization techniques, the training process is monitored in real time. This solves the efficiency bottleneck and real-time insufficiency of device performance evaluation in large-scale cloud training, and achieves real-time integrity verification and security assurance of the model training process.

CN121436093APending Publication Date: 2026-01-30ZHEJIANG UNIV OF TECH
0 Cites 0 Cited by

Patent Information

Application Number
CN202511275508.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-09-08
Publication Date
2026-01-30

AI Technical Summary

Technical Problem

Existing methods for verifying the integrity of model training suffer from a lack of performance evaluation mechanisms, efficiency bottlenecks, and insufficient real-time performance. This makes it difficult to effectively monitor the performance of training devices during large-scale training in the cloud, resulting in wasted computing resources and economic losses.

Method used

By constructing an operator runtime prediction model, combining a Trusted Execution Environment (TEE) and a cryptographic hash function, the training process is monitored in real time, the integrity of the model training is periodically verified, the device performance is evaluated using micro-baseline time and actual TEE timing, and quantization technology is used to tolerate floating-point precision differences, thereby achieving real-time integrity verification of the training process.

Benefits of technology

It enables real-time integrity verification of the model training process, avoiding waste of computing resources and degradation of model performance, reducing computing power waste and data risks, improving the robustness and scalability of the verification mechanism, and ensuring the correctness and security of the training process.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a model training integrity verification method based on a trusted execution environment. The method comprises the following steps: acquiring a model offline training data set; constructing an operator training set to train an operator operation time prediction model; obtaining the overall expected time of each training round of the corresponding to-be-trained model; quantifying model parameters after single-round training of each model in the GPU equipment to obtain a parameter group of the corresponding model; corresponding model training is carried out in the target GPU, TEE training and timing are started according to the pseudo-random seeds, and quantitative check points and complete check points are generated; and repeatedly verifying the integrity of model training in the TEE according to the quantitative check points and the complete check points, and regarding the model with the integrity as a qualified model. According to the method, abnormity can be found in time in the training process, waste of computing resources and reduction of model performance are avoided, the storage burden can be effectively relieved, robustness and expandability are improved, and correctness and safety of training are guaranteed.
Need to check novelty before this filing date? Find Prior Art