RB-Net: Training Highly Accurate and Efficient Binary Neural Networks With Reshaped Point-Wise Convolution and Balanced Activation

In this paper, we find that the conventional convolution operation becomes the bottleneck for extremely efficient binary neural networks (BNNs). To address this issue, we open up a new direction by introducing a reshaped point-wise convolution (RPC) to replace the conventional one to build BNNs. Spe...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on circuits and systems for video technology Jg. 32; H. 9; S. 6414 - 6424
Hauptverfasser: Liu, Chunlei, Ding, Wenrui, Chen, Peng, Zhuang, Bohan, Wang, Yufeng, Zhao, Yang, Zhang, Baochang, Han, Yuqi
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York IEEE 01.09.2022
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Accuracy
balanced activation
Binary neural network
Complexity
Computational efficiency
Computing costs
Convolution
Kernel
Neural networks
object classification
Remote procedure calls
reshaped point-wise convolution
Spatial data
Spatial databases
Training
ISSN:1051-8215, 1558-2205
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this paper, we find that the conventional convolution operation becomes the bottleneck for extremely efficient binary neural networks (BNNs). To address this issue, we open up a new direction by introducing a reshaped point-wise convolution (RPC) to replace the conventional one to build BNNs. Specifically, we conduct a point-wise convolution after rearranging the spatial information into depth, with which at least <inline-formula> <tex-math notation="LaTeX">2.25\times </tex-math></inline-formula> computation reduction can be achieved. Such an efficient RPC allows us to explore more powerful representational capacity of BNNs under a given computation complexity budget. Moreover, we propose to use a balanced activation (BA) to adjust the distribution of the scaled activations after binarization, which enables significant performance improvement of BNNs. After integrating RPC and BA, the proposed network, dubbed as RB-Net, strikes a good trade-off between accuracy and efficiency, achieving superior performance with lower computational cost against the state-of-the-art BNN methods. Specifically, our RB-Net achieves 66.8% Top-1 accuracy with ResNet-18 backbone on ImageNet, exceeding the state-of-the-art Real-to-Binary Net (65.4%) by 1.4% while achieving more than <inline-formula> <tex-math notation="LaTeX">3\times </tex-math></inline-formula> reduction (52M vs. 165M) in computational complexity.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1051-8215
1558-2205
DOI:10.1109/TCSVT.2022.3166803