Structure-Aware Automatic Channel Pruning by Searching with Graph Embedding
By: Zifan Liu , Yuan Cao , Yanwei Yu and more
Potential Business Impact:
Makes computer programs run faster and smaller.
Channel pruning is a powerful technique to reduce the computational overhead of deep neural networks, enabling efficient deployment on resource-constrained devices. However, existing pruning methods often rely on local heuristics or weight-based criteria that fail to capture global structural dependencies within the network, leading to suboptimal pruning decisions and degraded model performance. To address these limitations, we propose a novel structure-aware automatic channel pruning (SACP) framework that utilizes graph convolutional networks (GCNs) to model the network topology and learn the global importance of each channel. By encoding structural relationships within the network, our approach implements topology-aware pruning and this pruning is fully automated, reducing the need for human intervention. We restrict the pruning rate combinations to a specific space, where the number of combinations can be dynamically adjusted, and use a search-based approach to determine the optimal pruning rate combinations. Extensive experiments on benchmark datasets (CIFAR-10, ImageNet) with various models (ResNet, VGG16) demonstrate that SACP outperforms state-of-the-art pruning methods on compression efficiency and competitive on accuracy retention.
Similar Papers
Resource-Aware Neural Network Pruning Using Graph-based Reinforcement Learning
Machine Learning (CS)
Makes computer brains smaller, faster, and smarter.
C-SWAP: Explainability-Aware Structured Pruning for Efficient Neural Networks Compression
CV and Pattern Recognition
Makes computer "brains" smaller without losing smarts.
SepPrune: Structured Pruning for Efficient Deep Speech Separation
Sound
Makes AI hear clear voices in noisy places.