Filters








2,239 Hits in 6.1 sec

Partial Gated Feedback Recurrent Neural Network for Data Compression Type Classification

Hyewon Song, Beom Kwon, Hoon Yoo, Sanghoon Lee
2020 IEEE Access  
INDEX TERMS Compression type classification, deep learning, gated recurrent unit, lossless compression, partial gated feedback recurrent neural network, recurrent neural network.  ...  We emphasize on the temporal features, which consider a wide range of data, and spatial features from fully-connected layers to extract the feature vectors of each compression type.  ...  Recently, a convolutional neural network (CNN)-based method of distinguishing dictionary-based compression algorithms has been proposed [9] .  ... 
doi:10.1109/access.2020.3015493 fatcat:3ailoniwlnfufhnocekilmufaa

An Intelligent EEG Classification Methodology Based on Sparse Representation Enhanced Deep Learning Networks

Jing-Shan Huang, Yang Li, Bin-Qiang Chen, Chuang Lin, Bin Yao
2020 Frontiers in Neuroscience  
Aiming to achieve intelligent classification of EEG types with high accuracy, a classification methodology using sparse representation (SR) and fast compression residual convolutional neural networks (  ...  The common spatial patterns algorithm is used to obtain the features of the EEG signal. Subsequently, the redundant dictionary with sparse representation is constructed based on these features.  ...  Finally, EEG types are classified by the fast compression residual CNNs (FCRes-CNNs) classifier intelligently.  ... 
doi:10.3389/fnins.2020.00808 pmid:33177970 pmcid:PMC7596898 fatcat:ic5fnvfoknb3lhgiabysldk3h4

Large Scale Product Categorization using Structured and Unstructured Attributes [article]

Abhinandan Krishnan, Abilash Amarthaluri
2019 arXiv   pre-print
Product categorization using text data for eCommerce is a very challenging extreme classification problem with several thousands of classes and several millions of products to classify.  ...  Even though multi-class text classification is a well studied problem both in academia and industry, most approaches either deal with treating product content as a single pile of text, or only consider  ...  Figure a. shows the Multi-CNN architecture and Figure b. shows the Multi-LSTM architecture.  ... 
arXiv:1903.04254v1 fatcat:sle4le4rwrcqvmcymwy6svqrou

LCNN: Lookup-based Convolutional Neural Network [article]

Hessam Bagherinezhad, Mohammad Rastegari, Ali Farhadi
2017 arXiv   pre-print
We introduce LCNN, a lookup-based convolutional neural network that encodes convolutions by few lookups to a dictionary that is trained to cover the space of weights in CNNs.  ...  Our experimental results on ImageNet challenge show that LCNN can offer 3.2x speedup while achieving 55.1% top-1 accuracy using AlexNet architecture.  ...  A dictionary D ∈ R m×k can be used in any convolutional layer with input channel size m in any CNN architecture.  ... 
arXiv:1611.06473v2 fatcat:mztjabqn2bhbnlrjrlw6rnwkva

LCNN: Lookup-Based Convolutional Neural Network

Hessam Bagherinezhad, Mohammad Rastegari, Ali Farhadi
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We introduce LCNN, a lookup-based convolutional neural network that encodes convolutions by few lookups to a dictionary that is trained to cover the space of weights in CNNs.  ...  Our experimental results on ImageNet challenge show that LCNN can offer 3.2× speedup while achieving 55.1% top-1 accuracy using AlexNet architecture.  ...  A dictionary D ∈ R m×k can be used in any convolutional layer with input channel size m in any CNN architecture.  ... 
doi:10.1109/cvpr.2017.98 dblp:conf/cvpr/BagherinezhadRF17 fatcat:3rbk5fwiu5aurpjwvvy5ntckzq

Power Quality Data Compression and Disturbances Recognition Based on Deep CS-BiLSTM Algorithm With Cloud-Edge Collaboration

Xin Xia, Chuanliang He, Yingjie Lv, Bo Zhang, ShouZhi Wang, Chen Chen, Haipeng Chen
2022 Frontiers in Energy Research  
This paper proposes a hybrid model based on distributed compressive sensing and a bi-directional long-short memory network to classify power quality disturbances.  ...  For power disturbance identification, a new network structure is designed to improve the classification accuracy and reduce the training time, and the training parameters are optimized using the Deep Deterministic  ...  Storage Based on Cloud Edge Collaboration Under the cloud-edge collaborative architecture (Ning et al., 2021) , the DCS-OMP edge algorithm is used to compress and collect the power quality data of s  ... 
doi:10.3389/fenrg.2022.874351 fatcat:eub5sunpnjgmdpnmpiant5eu2m

Intelligent diagnosis with Chinese electronic medical records based on convolutional neural networks

Xiaozheng Li, Huazhen Wang, Huixin He, Jixiang Du, Jian Chen, Jinzhun Wu
2019 BMC Bioinformatics  
The novelty of this paper is reflected in the following: (1) We construct a pediatric medical dictionary based on Chinese EMRs. (2) Word2vec adopted in word embedding is used to achieve the semantic description  ...  Results: In this paper, we propose a deep learning framework to study intelligent diagnosis using Chinese EMR data, which incorporates a convolutional neural network (CNN) into an EMR classification application  ...  Based on the best CNN model architecture (onelayer CNN), the other classificaion applications, i.e., eight-classification application, 32-classification application, and 63-classification application,  ... 
doi:10.1186/s12859-019-2617-8 fatcat:vsddtl6yhrhrhh34u33kqsndra

Less is More: Matched Wavelet Pooling-Based Light-Weight CNNs with Application to Image Classification

Said EL-Khamy, Ahmad Al-Kabbany, Shimaa El-Bana
2022 IEEE Access  
), for light-weight CNN architectures, and in particular, MobileNets.  ...  A vast literature has focused on enhancing the performance of deep convolution models (CNNs) by introducing changes in the model architecture and/or developing new training procedures.  ...  A comparison between the classification performance metrics attained by wavelet pooling-based Mobileets using various wavelet types on STL-10. Please see text for more details. 4.  ... 
doi:10.1109/access.2022.3180498 fatcat:5tf53ddvv5agzllrv7qefus76u

Low-complexity CNNs for Acoustic Scene Classification [article]

Arshdeep Singh, Mark D. Plumbley
2022 arXiv   pre-print
We propose a low-complexity CNN architecture, and apply pruning and quantization to further reduce the parameters and memory.  ...  This paper presents a low-complexity framework for acoustic scene classification (ASC).  ...  For the purpose of open access, the authors have applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising. Thanks to James A.  ... 
arXiv:2207.11529v1 fatcat:gjlwr7jwubah7n6luo7nsqrp3u

Scenarios: A New Representation for Complex Scene Understanding [article]

Zachary A. Daniels, Dimitris N. Metaxas
2018 arXiv   pre-print
Training a single ScenarioNet model enables us to perform scene classification, scenario recognition, multi-object recognition, content-based scene image retrieval, and content-based image comparison.  ...  We learn scenarios from data using a novel matrix factorization method which we integrate into a new neural network architecture, the ScenarioNet.  ...  We use VGG-16 as our base CNN architecture, replacing the final fully-connected layers with the scenario block.  ... 
arXiv:1802.06117v1 fatcat:skakwkagmjdr5huqffwus6r2g4

Slim and Efficient Neural Network Design for Resource-Constrained SAR Target Recognition

Hongyi Chen, Fan Zhang, Bo Tang, Qiang Yin, Xian Sun
2018 Remote Sensing  
between compression ratio and classification accuracy.  ...  First, we introduce a new weight-based network pruning and adaptive architecture squeezing method to reduce the network storage and the time of inference and training process, meanwhile maintain a balance  ...  However, the problem of identifying a slimmer CNN architecture is challenging.  ... 
doi:10.3390/rs10101618 fatcat:lfrmlwujarhqpds3nj65zozcri

Deep Blind Compressed Sensing [article]

Shikha Singh, Vanika Singhal, Angshul Majumdar
2016 arXiv   pre-print
This work extends the recently proposed framework of deep matrix factorization in combination with blind compressed sensing; hence the term deep blind compressed sensing.  ...  In all cases, the superiority of our proposed deep blind compressed sensing can be envisaged.  ...  Based on the neural network type interpretation of dictionary learning [12] proposed deeper architectures using dictionary learning as the basic building block.  ... 
arXiv:1612.07453v1 fatcat:xcctnv4bbfbbfmfmuotrs4lu6i

Automatic Target Recognition on Synthetic Aperture Radar Imagery: A Survey [article]

O. Kechagias-Stamatis, N. Aouf
2020 arXiv   pre-print
Based on the current methodology trends, we propose a taxonomy for the SAR ATR architectures, along with a direct comparison of the strengths and weaknesses of each method under both standard and extended  ...  Spurred by this and given that Synthetic Aperture Radar (SAR) presents several advantages over its counterpart data domains, this paper surveys and assesses current SAR ATR architectures that employ the  ...  CNN-based architectures incorporate various schemes ranging from manual or GAN-based data augmentation to CNN distillation strategies, transfer learning and fully exploiting pretrained state-of-the-art  ... 
arXiv:2007.02106v2 fatcat:qz3kfu5varasjkagj44pw5byei

Fusing Deep Learning and Sparse Coding for SAR ATR

Odysseas Kechagias-Stamatis, Nabil Aouf
2018 IEEE Transactions on Aerospace and Electronic Systems  
Our architecture fuses a proposed Clustered version of the AlexNet Convolutional Neural Network with Sparse Coding theory that is extended to facilitate an adaptive elastic net optimization concept.  ...  We propose a multi-modal and multi-discipline data fusion strategy appropriate for Automatic Target Recognition (ATR) on Synthetic Aperture Radar imagery.  ...  Sparse Representation Classification (SRC) or Sparse Coding (SC) type of solutions aim at recovering the SAR testing imagery out of a dictionary where the SAR training images are the dictionary's base  ... 
doi:10.1109/taes.2018.2864809 fatcat:lsn5aiopobfvtoohy6ybasooby

Data-driven geophysics: from dictionary learning to deep learning [article]

Siwei Yu, Jianwei Ma
2020 arXiv   pre-print
In this article, we review the basic concepts of and recent advances in data-driven approaches from dictionary learning to deep learning in a variety of geophysical scenarios.  ...  We present a coding tutorial and a summary of tips for beginners and interested geophysical readers to rapidly explore deep learning.  ...  Acknowledgments The work was supported in part by the National Key Research and Development Program Data Availability Statement Data were not used, nor created for this research.  ... 
arXiv:2007.06183v2 fatcat:ow45ejo7izbkpmssedwd74rbym
« Previous Showing results 1 — 15 out of 2,239 results