Mixed Pooling for Convolutional Neural Networks

Yu, Dingjun; Wang, Hanli; Chen, Peiqiu; Wei, Zhihua

doi:10.1007/978-3-319-11740-9_34

Cited by 336 publications

(179 citation statements)

References 8 publications

Supporting

Mentioning

154

Contrasting

Order By: Relevance

“…However, alternative forms of pooling exist, i.e. mixed [24], stochastic [23], etc. that enhance the strengths and improve upon the weaknesses of average and max pooling.…”

Section: Pooling Layermentioning

confidence: 99%

An Ensemble of Convolutional Neural Networks Using Wavelets for Image Classification

Williams¹,

Li²

2018

JSEA

View full text Add to dashboard Cite

Machine learning is an integral technology many people utilize in all areas of human life. It is pervasive in modern living worldwide, and has multiple usages. One application is image classification, embraced across many spheres of influence such as business, finance, medicine, etc. to enhance produces, causes, efficiency, etc. This need for more accurate, detail-oriented classification increases the need for modifications, adaptations, and innovations to Deep Learning Algorithms. This article used Convolutional Neural Networks (CNN) to classify scenes in the CIFAR-10 database, and detect emotions in the KDEF database. The proposed method converted the data to the wavelet domain to attain greater accuracy and comparable efficiency to the spatial domain processing. By dividing image data into subbands, important feature learning occurred over differing low to high frequencies. The combination of the learned low and high frequency features, and processing the fused feature mapping resulted in an advance in the detection accuracy. Comparing the proposed methods to spatial domain CNN and Stacked Denoising Autoencoder (SDA), experimental findings revealed a substantial increase in accuracy.

show abstract

“…However, alternative forms of pooling exist, i.e. mixed [24], stochastic [23], etc. that enhance the strengths and improve upon the weaknesses of average and max pooling.…”

Section: Pooling Layermentioning

confidence: 99%

An Ensemble of Convolutional Neural Networks Using Wavelets for Image Classification

Williams¹,

Li²

2018

JSEA

View full text Add to dashboard Cite

show abstract

“…For average pooling, the average of the pixel values captured by the pooling window is obtained during subsampling operations. It has been established in several research works that the type of pooling operation implemented in the subsampling layer affects network performance, especially for pattern invariance learning [14].…”

Section: Kernels Convolution and Subsamplingmentioning

confidence: 99%

Iris nevus diagnosis: convolutional neural network and deep belief network

Oyedotun¹,

Khashman²

2017

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

This work presents the diagnosis of iris nevus using a convolutional neural network (CNN) and deep belief network (DBN). Iris nevus is a pigmented growth (tumor) found in the front of the eye or around the pupil. It is seen that racial and environmental factors affect the iris color (e.g., blue, hazel, brown) of patients; hence, pigmented growths may be masked in the eye background or iris. In this work, some image processing techniques are applied to images to reinforce areas of interests in them, after which the considered classifiers are trained. We describe the automated diagnosis of iris nevus using neural network-based systems for the classification of eye images as "nevus affected" and "unaffected". Recognition rates of 93.35% and 93.67% were achieved for the CNN and DBN, respectively. Hence, the systems described in this work can be used satisfactorily for diagnosis or to reinforce the confidence in manual-visual diagnosis by medical experts.

show abstract

“…Thus, improving the subspace pooling ability is helpful when combining with the model-averaging technique of dropout. Spatial mixed pooling is another pooling method used in the subsampling stage [12]. In [12], the authors replaced the deterministic pooling operations with a stochastic procedure that weights the conventional max pooling and average pooling values by a Bernoulli distribution.…”

Section: Motivationmentioning

confidence: 99%

“…Spatial mixed pooling is another pooling method used in the subsampling stage [12]. In [12], the authors replaced the deterministic pooling operations with a stochastic procedure that weights the conventional max pooling and average pooling values by a Bernoulli distribution. The processing flow of spatial mixed pooling in the subsampling stage is given in Fig 3A. The 4 sub-blocks in the red dotted boxes in Fig 2A are taken as the inputs.…”

Section: Motivationmentioning

confidence: 99%

See 1 more Smart Citation

Improving deep convolutional neural networks with mixed maxout units

Zhao

Fu-xian

2017

PLoS ONE

View full text Add to dashboard Cite

Motivated by insights from the maxout-units-based deep Convolutional Neural Network (CNN) that "non-maximal features are unable to deliver" and "feature mapping subspace pooling is insufficient," we present a novel mixed variant of the recently introduced maxout unit called a mixout unit. Specifically, we do so by calculating the exponential probabilities of feature mappings gained by applying different convolutional transformations over the same input and then calculating the expected values according to their exponential probabilities. Moreover, we introduce the Bernoulli distribution to balance the maximum values with the expected values of the feature mappings subspace. Finally, we design a simple model to verify the pooling ability of mixout units and a Mixout-units-based Network-in-Network (NiN) model to analyze the feature learning ability of the mixout models. We argue that our proposed units improve the pooling ability and that mixout models can achieve better feature learning and classification performance.

show abstract

Mixed Pooling for Convolutional Neural Networks

Cited by 336 publications

References 8 publications

An Ensemble of Convolutional Neural Networks Using Wavelets for Image Classification

An Ensemble of Convolutional Neural Networks Using Wavelets for Image Classification

Iris nevus diagnosis: convolutional neural network and deep belief network

Improving deep convolutional neural networks with mixed maxout units

Contact Info

Product

Resources

About