Learning Convolutional Neural Networks in the Frequency Domain

• URL: http://arxiv.org/abs/2204.06718v2
• Date: Fri, 15 Apr 2022 10:10:30 GMT
• Title: Learning Convolutional Neural Networks in the Frequency Domain
• Authors: Hengyue Pan and Yixin Chen and Xin Niu and Wenbo Zhou
• Abstract summary: We propose a novel neural network model, namely CEMNet, that can be trained in frequency domain. We introduce Weight Fixation Mechanism to alleviate over-fitting, and analyze the working behavior of Batch Normalization, Leaky ReLU and Dropout. Experimental results imply that CEMNet works well in frequency domain, and achieve good performance on MNIST and CIFAR-10 databases.
• Score: 33.902889724984746
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Convolutional neural network (CNN) achieves impressive success in the field of computer vision during the past few decades. As the core of CNNs, image convolution operation helps CNNs to achieve good performance on image-related tasks. However, image convolution is hard to be implemented and parallelized. In this paper, we propose a novel neural network model, namely CEMNet, that can be trained in frequency domain. The most important motivation of this research is that we can use the very simple element-wise multiplication operation to replace the image convolution in frequency domain based on Cross-Correlation Theorem. We further introduce Weight Fixation Mechanism to alleviate over-fitting, and analyze the working behavior of Batch Normalization, Leaky ReLU and Dropout in frequency domain to design their counterparts for CEMNet. Also, to deal with complex inputs brought by DFT, we design two branch network structure for CEMNet. Experimental results imply that CEMNet works well in frequency domain, and achieve good performance on MNIST and CIFAR-10 databases. To our knowledge, CEMNet is the first model trained in Fourier Domain that achieves more than 70\% validation accuracy on CIFAR-10 database.

Related papers

• Auto-Train-Once: Controller Network Guided Automatic Network Pruning from Scratch [72.26822499434446]
  Auto-Train-Once (ATO) is an innovative network pruning algorithm designed to automatically reduce the computational and storage costs of DNNs. We provide a comprehensive convergence analysis as well as extensive experiments, and the results show that our approach achieves state-of-the-art performance across various model architectures.
  arXiv  Detail & Related papers  (2024-03-21T02:33:37Z)
• Multiscale Low-Frequency Memory Network for Improved Feature Extraction in Convolutional Neural Networks [13.815116154370834]
  We introduce a novel framework, the Multiscale Low-Frequency Memory (MLFM) Network. The MLFM efficiently preserves low-frequency information, enhancing performance in targeted computer vision tasks. Our work builds upon the existing CNN foundations and paves the way for future advancements in computer vision.
  arXiv  Detail & Related papers  (2024-03-13T00:48:41Z)
• TFDMNet: A Novel Network Structure Combines the Time Domain and Frequency Domain Features [34.91485245048524]
  This paper proposes a novel Element-wise Multiplication Layer (EML) to replace convolution layers. We also introduce a Weight Fixation mechanism to alleviate the problem of over-fitting. Experimental results imply that TFDMNet achieves good performance on MNIST, CIFAR-10 and ImageNet databases.
  arXiv  Detail & Related papers  (2024-01-29T08:18:21Z)
• Training Convolutional Neural Networks with the Forward-Forward algorithm [1.74440662023704]
  Forward Forward (FF) algorithm has up to now only been used in fully connected networks. We show how the FF paradigm can be extended to CNNs. Our FF-trained CNN, featuring a novel spatially-extended labeling technique, achieves a classification accuracy of 99.16% on the MNIST hand-written digits dataset.
  arXiv  Detail & Related papers  (2023-12-22T18:56:35Z)
• GMConv: Modulating Effective Receptive Fields for Convolutional Kernels [52.50351140755224]
  In convolutional neural networks, the convolutions are performed using a square kernel with a fixed N $times$ N receptive field (RF) Inspired by the property that ERFs typically exhibit a Gaussian distribution, we propose a Gaussian Mask convolutional kernel (GMConv) in this work. Our GMConv can directly replace the standard convolutions in existing CNNs and can be easily trained end-to-end by standard back-propagation.
  arXiv  Detail & Related papers  (2023-02-09T10:17:17Z)
• InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions [95.94629864981091]
  This work presents a new large-scale CNN-based foundation model, termed InternImage, which can obtain the gain from increasing parameters and training data like ViTs. The proposed InternImage reduces the strict inductive bias of traditional CNNs and makes it possible to learn stronger and more robust patterns with large-scale parameters from massive data like ViTs.
  arXiv  Detail & Related papers  (2022-11-10T18:59:04Z)
• Global Filter Networks for Image Classification [90.81352483076323]
  We present a conceptually simple yet computationally efficient architecture that learns long-term spatial dependencies in the frequency domain with log-linear complexity. Our results demonstrate that GFNet can be a very competitive alternative to transformer-style models and CNNs in efficiency, generalization ability and robustness.
  arXiv  Detail & Related papers  (2021-07-01T17:58:16Z)
• Learning Frequency-aware Dynamic Network for Efficient Super-Resolution [56.98668484450857]
  This paper explores a novel frequency-aware dynamic network for dividing the input into multiple parts according to its coefficients in the discrete cosine transform (DCT) domain. In practice, the high-frequency part will be processed using expensive operations and the lower-frequency part is assigned with cheap operations to relieve the computation burden. Experiments conducted on benchmark SISR models and datasets show that the frequency-aware dynamic network can be employed for various SISR neural architectures.
  arXiv  Detail & Related papers  (2021-03-15T12:54:26Z)
• IC Networks: Remodeling the Basic Unit for Convolutional Neural Networks [8.218732270970381]
  "Inter-layer Collision" (IC) structure can be integrated into existing CNNs to improve their performance. New training method, namely weak logit distillation (WLD), is proposed to speed up the training of IC networks. In the ImageNet experiment, we integrate the IC structure into ResNet-50 and reduce the top-1 error from 22.38% to 21.75%.
  arXiv  Detail & Related papers  (2021-02-06T03:15:43Z)
• Frequency learning for image classification [1.9336815376402716]
  This paper presents a new approach for exploring the Fourier transform of the input images, which is composed of trainable frequency filters. We propose a slicing procedure to allow the network to learn both global and local features from the frequency-domain representations of the image blocks.
  arXiv  Detail & Related papers  (2020-06-28T00:32:47Z)
• Binarizing MobileNet via Evolution-based Searching [66.94247681870125]
  We propose a use of evolutionary search to facilitate the construction and training scheme when binarizing MobileNet. Inspired by one-shot architecture search frameworks, we manipulate the idea of group convolution to design efficient 1-Bit Convolutional Neural Networks (CNNs) Our objective is to come up with a tiny yet efficient binary neural architecture by exploring the best candidates of the group convolution.
  arXiv  Detail & Related papers  (2020-05-13T13:25:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.