Enhanced Pruning Strategy for Multi-Component Neural Architectures Using Component-Aware Graph Analysis
This work addresses the deployment of complex neural networks in resource-limited environments, but it appears incremental as it extends existing pruning frameworks with component-specific modifications.
The paper tackles the problem of pruning Multi-Component Neural Architectures (MCNAs) to reduce model size for resource-constrained settings, introducing a component-aware pruning strategy that achieves greater sparsity and reduced performance degradation in a control task.
Deep neural networks (DNNs) deliver outstanding performance, but their complexity often prohibits deployment in resource-constrained settings. Comprehensive structured pruning frameworks based on parameter dependency analysis reduce model size with specific regard to computational performance. When applying them to Multi-Component Neural Architectures (MCNAs), they risk network integrity by removing large parameter groups. We introduce a component-aware pruning strategy, extending dependency graphs to isolate individual components and inter-component flows. This creates smaller, targeted pruning groups that conserve functional integrity. Demonstrated effectively on a control task, our approach achieves greater sparsity and reduced performance degradation, opening a path for optimizing complex, multi-component DNNs efficiently.