A Novel Image Tag Completion Method Based on Convolutional Neural Transformation

Geng, Yanyan; Zhang, Guohui; Li, Weizhi; Gu, Yi; Liang, Ru‐Ze; Liang, Guoxi; Wang, Jingbin; Wu, Yongxiang; Patil, Nitin T.; Wang, Jingyan

doi:10.1007/978-3-319-68612-7_61

Cited by 43 publications

(13 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When training network convergence speed of the BP algorithm is not satisfactory, the thresholds and weights of the nodes at the hidden layers of the BP neural network would be taken as the input information of the genetic algorithm. And they are further coded to generate chromosomes [20]. The later generations generated with selection operators, crossover operators, and mutation operators of the genetic algorithm are then used to generate new descendants as initial values of the BP algorithm.…”

Section: Bp Neural Network Modelmentioning

confidence: 99%

Big Data Digging of the Public’s Cognition about Recycled Water Reuse Based on the BP Neural Network

Liu

Wang

et al. 2018

Complexity

View full text Add to dashboard Cite

Reuse of recycled water is very important to both the environment and economy, while the public cognition degree towards recycled water reuse also plays a key role in this process, and it determines the acceptance degree of the public towards recycled water reuse. Under the background of the big data, the Hadoop platform was used to collect and save data about the public’s cognition towards recycled water in one city and the BP neural network algorithm was used to construct an evaluation model that could affect the public’s cognition level. The public’s risk perception, subjective norm, and perceived behavioral control regarding recycled water reuse were selected as key factors. Based on a multivariate clustering algorithm, MATLAB software was used to make real testing on massive effective data and assumption models, so as to analyze the proportion of three evaluation factors and understand the simulation parameter scope of the cognition degree of different groups of citizens. Lastly, several suggestions were proposed to improve the public’s cognition on recycled water reuse based on the big data in terms of policy mechanism.

show abstract

Section: Bp Neural Network Modelmentioning

confidence: 99%

Big Data Digging of the Public’s Cognition about Recycled Water Reuse Based on the BP Neural Network

Liu

Wang

et al. 2018

Complexity

View full text Add to dashboard Cite

show abstract

“…Max-pooling Figure 2: Overview of the proposed deep learning framework. 4 Computational Intelligence and Neuroscience Problem optimization: it is difficult to solve the problem in (10) directly, because the classification map p i itself is a function of W, f, and p j: X j ∈N i according to (2) and 5:…”

Section: Conv Xmentioning

confidence: 99%

“…Background. Deep learning has been proven to be a powerful tool for pattern classification problems and sensor studies [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15]. A deep learning model usually has more than three layers, and by using multiple layers, the model extracts hierarchical features from the original data.…”

Section: Introductionmentioning

confidence: 99%

Iterative Deep Neighborhood: A Deep Learning Model Which Involves Both Input Data Points and Their Neighbors

Liu

Yan

et al. 2020

Computational Intelligence and Neuroscience

Self Cite

View full text Add to dashboard Cite

Deep learning models, such as deep convolutional neural network and deep long-short term memory model, have achieved great successes in many pattern classification applications over shadow machine learning models with hand-crafted features. The main reason is the ability of deep learning models to automatically extract hierarchical features from massive data by multiple layers of neurons. However, in many other situations, existing deep learning models still cannot gain satisfying results due to the limitation of the inputs of models. The existing deep learning models only take the data instances of an input point but completely ignore the other data points in the dataset, which potentially provides critical insight for the classification of the given input. To overcome this gap, in this paper, we show that the neighboring data points besides the input data point itself can boost the deep learning model’s performance significantly and design a novel deep learning model which takes both the data instances of an input point and its neighbors’ classification responses as inputs. In addition, we develop an iterative algorithm which updates the neighbors of data points according to the deep representations output by the deep learning model and the parameters of the deep learning model alternately. The proposed algorithm, named “Iterative Deep Neighborhood (IDN),” shows its advantages over the state-of-the-art deep learning models over tasks of image classification, text sentiment analysis, property price trend prediction, etc.

show abstract

“…In recent years, deep learning has been proven to be the most powerful data representation method (Chao, Zhi, Dong and Liu, 2018;Chu, Huang, Xie, Tan, Kamal and Xiong, 2018;Geng, Zhang, Li, Gu, Liang, Liang, Wang, Wu, Patil and Wang, 2017;Glorot, Bordes and Bengio, 2011;Guo, Liu, Oerlemans, Lao, Wu and Lew, 2016;Hu, Wang, Peng, Qiu, Shi and Liu, 2018;Längkvist, Karlsson and Loutfi, 2014;LeCun, Bengio and Hinton, 2015;Ngiam, Khosla, Kim, Nam, Lee and Ng, 2011;Sadouk, Gadi and Essoufi, 2018;Schmidhuber, 2015;Voulodimos, Doulamis, Bebis and Stathaki, 2018a;Voulodimos, Doulamis, Doulamis and Protopapadakis, 2018b;Wu, Zhai, Li, Cui, Wang and Patil;Zhang, Liang, Li, Fang, Wang, Geng and Wang, 2017;Zhang, Liang, Su, Qu and Wang, 2018a). Deep learning methods learn a neural network of multiple layers to extract the hierarchical patterns from the original data, and provide high-level and abstractive features for the learning problems.…”

Section: Backgroundsmentioning

confidence: 99%

Low-Rank Deep Convolutional Neural Network for Multitask Learning

Shang

Wang³

2019

Computational Intelligence and Neuroscience

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a novel multitask learning method based on the deep convolutional network. The proposed deep network has four convolutional layers, three max-pooling layers, and two parallel fully connected layers. To adjust the deep network to multitask learning problem, we propose to learn a low-rank deep network so that the relation among different tasks can be explored. We proposed to minimize the number of independent parameter rows of one fully connected layer to explore the relations among different tasks, which is measured by the nuclear norm of the parameter of one fully connected layer, and seek a low-rank parameter matrix. Meanwhile, we also propose to regularize another fully connected layer by sparsity penalty so that the useful features learned by the lower layers can be selected. The learning problem is solved by an iterative algorithm based on gradient descent and back-propagation algorithms. The proposed algorithm is evaluated over benchmark datasets of multiple face attribute prediction, multitask natural language processing, and joint economics index predictions. The evaluation results show the advantage of the low-rank deep CNN model over multitask problems.

show abstract

A Novel Image Tag Completion Method Based on Convolutional Neural Transformation

Cited by 43 publications

References 36 publications

Big Data Digging of the Public’s Cognition about Recycled Water Reuse Based on the BP Neural Network

Big Data Digging of the Public’s Cognition about Recycled Water Reuse Based on the BP Neural Network

Iterative Deep Neighborhood: A Deep Learning Model Which Involves Both Input Data Points and Their Neighbors

Low-Rank Deep Convolutional Neural Network for Multitask Learning

Contact Info

Product

Resources

About