提出可量化保证的神经网络压缩验证框架，提升安全系统可靠性。
SimCert: Probabilistic Certification for Behavioral Similarity in Deep Neural Network Compression
- 用符号传播法同时支持量化与剪枝的压缩验证
- 基于伯恩斯坦不等式的方差感知边界，提升验证精度
- 适合自动驾驶等安全关键系统的模型压缩验证
在资源受限的嵌入式系统中部署深度神经网络需采用量化和剪枝等压缩技术。然而，确保压缩模型保留原始设计的行为一致性，是安全关键系统设计中的核心挑战。现有验证方法往往缺乏可扩展性或无法处理剪枝带来的结构异质性。本文提出 SimCert，一种用于压缩神经网络行为相似性的概率认证框架。不同于最坏情况分析，SimCert 提供可调节置信度的定量安全保证。其核心包括：(1) 支持量化与剪枝的双网络符号传播方法；(2) 基于伯恩斯坦不等式的方差感知边界技术，以收紧安全证书；(3) 自动化验证工具链。在 ACAS Xu 及计算机视觉基准上的实验表明，SimCert 显著优于现有最先进方法。
原文摘要 · Abstract (English)
Deploying Deep Neural Networks (DNNs) on resource-constrained embedded systems requires aggressive model compression techniques like quantization and pruning. However, ensuring that the compressed model preserves the behavioral fidelity of the original design is a critical challenge in the safety-critical system design flow. Existing verification methods often lack scalability or fail to handle the architectural heterogeneity introduced by pruning. In this work, we propose SimCert, a probabilistic certification framework for verifying the behavioral similarity of compressed neural networks. Unlike worst-case analysis, SimCert provides quantitative safety guarantees with adjustable confidence levels. Our framework features: (1) A dual-network symbolic propagation method supporting both quantization and pruning; (2) A variance-aware bounding technique using Bernstein's inequality to tighten safety certificates; and (3) An automated verification toolchain. Experimental results on ACAS Xu and computer vision benchmarks demonstrate that SimCert outperforms state-of-the-art baselines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。