#6062. FLOPs-efficient filter pruning via transfer scale for neural network acceleration

July 2026publication date
Proposal available till 29-05-2025
4 total number of authors per manuscript0 $

The title of the journal is available only for the authors who have already paid for
Journal’s subject area:
Modeling and Simulation;
Computer Science (all);
Theoretical Computer Science;
Places in the authors’ list:
place 1place 2place 3place 4
FreeFreeFreeFree
2350 $1200 $1050 $900 $
Contract №6062.1 Contract №6062.2 Contract №6062.3 Contract №6062.4
1 place - free (for sale)
2 place - free (for sale)
3 place - free (for sale)
4 place - free (for sale)

Abstract:
Model pruning is a useful technique to reduce the computational cost of convolutional neural networks. In this paper, we first propose a simple but effective filter level pruning criterion, which assesses the importance of a filter by exploring the transfer scale (TS) of its feature maps in the next layer. The principle is that for a trained CNN model, an important filter should have strong connections with the next layer, otherwise the transfer scale of its feature map will be low and hence removing it will have little influence on the network. Besides, we observe that filters from the computationally-intensive layers are more sensitive to pruning, which makes it difficult to further compress the floating-point operations (FLOPs) of the model without reducing accuracy. To solve this problem, we propose a FLOPs-efficient group Lasso approach for TS to guide the network to use fewer filters in the computationally-intensive layers, which leads to better FLOPs compression performance after pruning. We refer to the proposed method as FETS. Compared with the state-of-the-art methods, our FETS achieves similar or better accuracy, but with significantly larger FLOPs compression ratio.
Keywords:
Machine learning; Network compression; Network pruning

Contacts :
0