SimCert: Probabilistic Certification for Behavioral Similarity in Deep Neural Network Compression
This addresses the challenge of verifying compressed models in safety-critical embedded systems, representing an incremental improvement over existing methods.
The paper tackles the problem of ensuring behavioral fidelity in compressed deep neural networks for safety-critical systems, proposing SimCert, a probabilistic certification framework that provides quantitative safety guarantees and outperforms state-of-the-art baselines on benchmarks like ACAS Xu and computer vision tasks.
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.