Enhanced Network Compression Through Tensor Decompositions and Pruning

Van Tien Pham

doi:10.1109/TNNLS.2024.3370294

Enhanced Network Compression Through Tensor Decompositions and Pruning

2025-03-01

1

OA

AI

DOI：10.1109/TNNLS.2024.3370294

Original

Original request for help

Save

Abstract

En 中文

Network compression techniques that combine tensor decompositions and pruning have shown promise in leveraging the advantages of both strategies. In this work, we propose enhanced Network cOmpRession through TensOr decompositions and pruNing (NORTON), a novel method for network compression. NORTON introduces the concept of filter decomposition, enabling a more detailed decomposition of the network while preserving the weight's multidimensional properties. Our method incorporates a novel structured pruning approach, effectively integrating the decomposed model. Through extensive experiments on various architectures, benchmark datasets, and representative vision tasks, we demonstrate the usefulness of our method. NORTON achieves superior results compared to state-of-the-art (SOTA) techniques in terms of complexity and accuracy. Our code is also available for research purposes.

Keywords:

Efficient inference

hybrid compression

structured pruning

tensor decompositions

Enhanced Network Compression Through Tensor Decompositions and Pruning

Abstract

Journal

Organization