Learning Frequency Domain Approximation for Binary Neural Networks

Xu, Yixing; Han, Kai; Xu, Chang; Xu, Chunjing; Wang, Yunhe

doi:10.48550/arxiv.2103.00841

Cited by 2 publications

(2 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SI-BNN [79] proposed trainable parameters for activations and gradients in the back-propagation. In addition, the Fourier Frequency Domain Approximation (FDA) is used to update the gradients in back-propagation in [80]. J. Lee et al [81] introduced an Element-Wise Gradient Scaling (EWGS) to update each gradient element by scaling factor.…”

Section: C: Gradient Error Minimizationmentioning

confidence: 99%

A Systematic Literature Review on Binary Neural Networks

et al. 2023

View full text Add to dashboard Cite

This paper presents an extensive literature review on Binary Neural Network (BNN). BNN utilizes binary weights and activation function parameters to substitute the full-precision values. In digital implementations, BNN replaces the complex calculations of Convolutional Neural Networks (CNNs) with simple bitwise operations. BNN optimizes large computation and memory storage requirements, which leads to less area and power consumption compared to full-precision models. Although there are many advantages of BNN, the binarization process has a significant impact on the performance and accuracy of the generated models. To reflect the state-of-the-art in BNN and explore how to develop and improve BNNbased models, we conduct a systematic literature review on BNN with data extracted from 239 research studies. Our review discusses various BNN architectures and the optimization approaches developed to improve their performance. There are three main research directions in BNN: accuracy optimization, compression optimization, and acceleration optimization. The accuracy optimization approaches include quantization error reduction, special regularization, gradient error minimization, and network structure. The compression optimization approaches combine fractional BNN and pruning. The acceleration optimization approaches comprise computing in-memory, FPGA-based implementations, and ASIC-based implementations. At the end of our review, we present a comprehensive analysis of BNN applications and their evaluation metrics. Also, we shed some light on the most common BNN challenges and the future research trends of BNN.

show abstract

Section: C: Gradient Error Minimizationmentioning

confidence: 99%

A Systematic Literature Review on Binary Neural Networks

et al. 2023

View full text Add to dashboard Cite

show abstract

“…General model compression approaches fall under multiple forms [12]: pruning [21,63], quantization [64,56,16], knowledge distillation [27,44], as well as their compositions [61,69,71]. A Binary Neural Network (BNN) [13,14,34,73,51,14,36,48,44,29,72,39,25,37,10,57,20,66] represents the most extreme form of model quantization as it quantizes weights in convolution layers to only 1 bit, enjoying great speed-up compared with its full-precision counterpart. [50] roughly divides previous BNN literature into two categories: (i) native BNN [13,14,34] which directly applies binarization to a full-precision model by a pre-defined binarization function.…”

Section: Related Workmentioning

confidence: 99%

"BNN - BN = ?": Training Binary Neural Networks without Batch Normalization

Chen¹,

Zhang²,

Ouyang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Batch normalization (BN) is a key facilitator and considered essential for state-of-the-art binary neural networks (BNN). However, the BN layer is costly to calculate and is typically implemented with non-binary parameters, leaving a hurdle for the efficient implementation of BNN training. It also introduces undesirable dependence between samples within each batch. Inspired by the latest advance on Batch Normalization Free (BN-Free) training [7], we extend their framework to training BNNs, and for the first time demonstrate that BNs can be completed removed from BNN training and inference regimes. By plugging in and customizing techniques including adaptive gradient clipping, scale weight standardization, and specialized bottleneck block, a BN-free BNN is capable of maintaining competitive accuracy compared to its BN-based counterpart. Extensive experiments validate the effectiveness of our proposal across diverse BNN backbones and datasets. For example, after removing BNs from the state-of-the-art ReActNets [38], it can still be trained with our proposed methodology to achieve 92.08%, 68.34%, and 68.0% accuracy on CIFAR-10, CIFAR-100, and ImageNet respectively, with marginal performance drop (0.23% ∼ 0.44% on CIFAR and 1.40% on ImageNet). Codes and pre-trained models are available at: https://github.com/VITA-Group/BNN_NoBN .

show abstract

Learning Frequency Domain Approximation for Binary Neural Networks

Cited by 2 publications

References 23 publications

A Systematic Literature Review on Binary Neural Networks

A Systematic Literature Review on Binary Neural Networks

"BNN - BN = ?": Training Binary Neural Networks without Batch Normalization

Contact Info

Product

Resources

About