FILTER PRUNING BASED ON LOCAL GRADIENT ACTIVATION MAPPING IN CONVOLUTIONAL NEURAL NETWORKS
| dc.contributor.author | Intraraprasit, Monthon | |
| dc.contributor.author | Chitsobhuk, Orachat | |
| dc.date.accessioned | 2026-08-06T10:42:54Z | |
| dc.date.available | 2026-08-06T10:42:54Z | |
| dc.date.issued | 2023-12-01 | |
| dc.description.abstract | Convolutional Neural Network (CNN) is a well-known Deep learning model utilized extensively in the field of computer vision. The structure of convolutional neural networks is quite complicated and necessitates a substantial amount of computational time and storage resources. As a result, it is difficult to adopt a CNN model on a resource-constraint device. Model pruning can help to reduce computation time and storage re-quirements. In this research, we propose a filter pruning technique based on Localized Gradient Activation heatmaP (LGAP) for the purpose of pruning CNNs. Analyzing a filter based on statistical criterion of single neuron can lead to a loss in spatial relations within the filter activation itself, the relationship to target prediction, as well as the relationship among filters in that specific layer. To minimize the limitations, we evaluate the significance of a filter through the spatial information of local gradient activation related to the target prediction in terms of the layer-wise loss of the investigated filter. The effect of loss of an investigated filter demonstrates the significance or insignificance of the filter. Our pruning criteria ensure that these significant filters are preserved, while maintaining the model accuracy. The performance of our pruning method was validated using VGG-16 and ResNet-50. With pruning ratio of 50%, VGG-16 tends to decrease 1.66% of its accuracy, 3.6× of FLOP and 3.9× of storage reduction. For ResNet-50, with 50% pruning ratio, the results show that Top-1 and Top-5 of our pruning techniques outperform all the baseline techniques with a reduction of top-1 accuracy by 3.56%, top-5 accuracy by 1.89%, Floating Point Operation by 2.3×, and storage by 2.05×. | |
| dc.identifier.citation | International Journal of Innovative Computing Information and Control, 19(6), 1697-1715, 2023 | |
| dc.identifier.doi | 10.24507/ijicic.19.06.1697 | |
| dc.identifier.issn | 13494198 | |
| dc.identifier.other | 2-s2.0-85172125937 | |
| dc.identifier.uri | https://dspace.kmitl.ac.th/handle/123456789/14783 | |
| dc.source | International Journal of Innovative Computing Information and Control | |
| dc.subject | Convolutional neural networks | |
| dc.subject | Deep learning | |
| dc.subject | Filter pruning | |
| dc.subject | Model compression | |
| dc.title | FILTER PRUNING BASED ON LOCAL GRADIENT ACTIVATION MAPPING IN CONVOLUTIONAL NEURAL NETWORKS | |
| dc.type | Article |