Saved in:
Bibliographic Details
Main Authors: Li, Jingyang, Li, Guoqiang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.04268
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915326988386304
author Li, Jingyang
Li, Guoqiang
author_facet Li, Jingyang
Li, Guoqiang
contents The rapid development of deep learning has led to challenges in deploying neural networks on edge devices, mainly due to their high memory and runtime complexity. Network compression techniques, such as quantization and pruning, aim to reduce this complexity while maintaining accuracy. However, existing incremental verification methods often focus only on quantization and struggle with structural changes. This paper presents MUC-G4 (Minimal Unsat Core-Guided Incremental Verification), a novel framework for incremental verification of compressed deep neural networks. It encodes both the original and compressed networks into SMT formulas, classifies changes, and use \emph{Minimal Unsat Cores (MUCs)} from the original network to guide efficient verification for the compressed network. Experimental results show its effectiveness in handling quantization and pruning, with high proof reuse rates and significant speedup in verification time compared to traditional methods. MUC-G4 hence offers a promising solution for ensuring the safety and reliability of compressed neural networks in practical applications.
format Preprint
id arxiv_https___arxiv_org_abs_2506_04268
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MUC-G4: Minimal Unsat Core-Guided Incremental Verification for Deep Neural Network Compression
Li, Jingyang
Li, Guoqiang
Machine Learning
Artificial Intelligence
The rapid development of deep learning has led to challenges in deploying neural networks on edge devices, mainly due to their high memory and runtime complexity. Network compression techniques, such as quantization and pruning, aim to reduce this complexity while maintaining accuracy. However, existing incremental verification methods often focus only on quantization and struggle with structural changes. This paper presents MUC-G4 (Minimal Unsat Core-Guided Incremental Verification), a novel framework for incremental verification of compressed deep neural networks. It encodes both the original and compressed networks into SMT formulas, classifies changes, and use \emph{Minimal Unsat Cores (MUCs)} from the original network to guide efficient verification for the compressed network. Experimental results show its effectiveness in handling quantization and pruning, with high proof reuse rates and significant speedup in verification time compared to traditional methods. MUC-G4 hence offers a promising solution for ensuring the safety and reliability of compressed neural networks in practical applications.
title MUC-G4: Minimal Unsat Core-Guided Incremental Verification for Deep Neural Network Compression
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.04268