A Hybrid Quantization Method for Image Recognition Models Based on Bit Flip Attack
By optimizing the objective function through bit-flipping attacks and evolutionary algorithms, the optimal mixed precision quantization strategy for image recognition models is determined, solving the problem of high model storage space and computing resources in existing technologies, and achieving rapid mixed quantization and improved recognition speed.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JILIN UNIVERSITY
- Filing Date
- 2023-12-18
- Publication Date
- 2026-05-26
AI Technical Summary
Existing technologies struggle to rapidly achieve hybrid quantization of deep learning models while maintaining image recognition accuracy, and they also require excessive computing resources and storage space.
The sensitivity of model parameters is recorded layer by layer using a bit-flipping attack method. The objective function is optimized by combining the bit-flipping attack with the evolutionary algorithm to determine the optimal mixed precision quantization strategy. The optimal bit width is set for different layers of the model. The optimal mixed precision quantization strategy is searched through bit-flipping attack and evolutionary algorithm to optimize the storage and recognition speed of the model.
It significantly reduces the model's storage space requirements and improves recognition speed while maintaining image recognition accuracy.
Smart Images

Figure CN117593631B_ABST