753 Hits in 2.6 sec

Multipath Graph Convolutional Neural Networks [article]

Rangan Das, Bikram Boote, Saumik Bhattacharya, Ujjwal Maulik
2021 arXiv   pre-print
In this work, we propose a novel Multipath Graph convolutional neural network that aggregates the output of multiple different shallow networks.  ...  Recent research has focused on stacking multiple layers like in convolutional neural networks for the increased expressive power of graph convolution networks.  ...  Convolution operation in GCNs is a generalization of the convolution operation used in convolution neural networks (CNNs) (Kipf and Welling 2016) .  ... 
arXiv:2105.01510v1 fatcat:mey2t3k72vegpg44foywvckhqm

ISCC 2020 Keyword Index

2020 2020 IEEE Symposium on Computers and Communications (ISCC)  
neural network Convolutional Neural Network convolutional neural networks Convolutional Neural Networks Cooperative Management Coronavirus Cortex-A9 MPCore Cost function modification Cost-Efficiency  ...  gas consumption Genetic algorithm Genetic Algorithm Glaucoma Gossip GPU Grammatical Evolution Graph Convolutional Network Graph Neural Networks grey model group key management guaranteed based  ... 
doi:10.1109/iscc50000.2020.9219679 fatcat:al6gjafwwneo5g5paprquzp7n4

GCLR: GNN based Cross Layer Optimization for Multipath TCP by Routing

Ting Zhu, Xiaohui Chen, Li Chen, Weidong Wang, Guo Wei
2020 IEEE Access  
INDEX TERMS Routing, multipath TCP, graph neural network, cross layer optimization, software defined networking. 17060 This work is licensed under a Creative Commons Attribution 4.0 License.  ...  To address these problems, in this paper, firstly, a novel Graph Neural Network (GNN) based multipath routing model is proposed to explore the complications among links, paths, subflows and the MPTCP connection  ...  Under the impetus of deep learning technologies, researchers propose the graph neural network [43] , which combines the thoughts of convolutional networks, cyclic networks, and deep auto-encoders.  ... 
doi:10.1109/access.2020.2966045 fatcat:ou4wkwyiqzeq7j5ionikghfqga

MRI based genomic analysis of glioma using three pathway deep convolutional neural network for IDH classification

2021 Turkish Journal of Electrical Engineering and Computer Sciences  
A 3-pathway convolutional neural network was trained for IDH 9 classification.  ...  The 13 results have demonstrated the multipath convolutional neural networks as state-of-the-art method with simple design to 14 predict IDH genotypes in glioma with auto-extraction of radiogenomic features  ...  Our deep neural network-based work proposes multipath convolutional neural 29 network with the capability to auto-discriminate IDH types of glioma.  ... 
doi:10.3906/elk-2104-180 fatcat:gzolynp6vfgu3gtel6sg5bsxae


2020 2020 13th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)  
Using Graph Theory Jing Ai, Tiantian Liu, Kexin Wang, Jian Zhang, Tianlin Huang ... ....407 Sentence Modeling via Graph Construction and Graph Neural Networks for Semantic Textual Similarity Ke Zhou  ...  ...........................................................231 Research on defect detection system of cloth based on convolutional neural network Zhang Qiyan, Li Mingjing, Yan Denghao, Yang Longbiao, Yu  ... 
doi:10.1109/cisp-bmei51763.2020.9263536 fatcat:7ulpvhnt35d2lg5dwzu4kexley

Automated Segmentation of Cervical Nuclei in Pap Smear Images using Deformable Multi-path Ensemble Model [article]

Jie Zhao, Quanzheng Li, Xiang Li, Hongfeng Li, Li Zhang
2019 arXiv   pre-print
The approach adopts a U-shaped convolutional network as a backbone network, in which dense blocks are used to transfer feature information more effectively.  ...  In this work, a method of automated cervical nuclei segmentation using Deformable Multipath Ensemble Model (D-MEM) is proposed.  ...  In order to solve the aforementioned problems, we propose segmenting cervical nuclei via Deformable Multipath Ensemble Model (D-MEM) based on novel deep neural networks.  ... 
arXiv:1812.00527v2 fatcat:gy6fdq3dlndjlhbckrckvko57q

A Deep Learning Approach to Position Estimation from Channel Impulse Responses

Arne Niitsoo, Thorsten Edelhäußer, Ernst Eberlein, Niels Hadaschik, Christopher Mutschler
2019 Sensors  
However, in industrial environments where multipath propagation is predominant it is difficult to extract the correct ToF of the signal.  ...  Our experiments show that our DL-based position estimation not only works well under harsh multipath propagation but also outperforms state-of-the-art approaches in line-of-sight situations.  ...  Convolutional neural networks (CNN) define a special architecture of neural networks.  ... 
doi:10.3390/s19051064 fatcat:i5md3oe5ijfg5ok5ss7cj5mthy

3D imaging from multipath temporal echoes [article]

Alex Turpin, Valentin Kapitany, Jack Radford, Davide Rovelli, Ashley Lyons, Ilya Starshynov, Daniele Faccio
2020 arXiv   pre-print
Numerical modelling and an information theoretic perspective prove the concept and provide insight into the role of the multipath information.  ...  Multipath sensing has also been combined with Bayesian inference [23] and convolutional neural networks [24] to localise sonic sources.  ...  neural network random.  ... 
arXiv:2011.09284v1 fatcat:dug3hnuaerdzza66i36ezzfjda

Neural RF SLAM for unsupervised positioning and mapping with channel state information [article]

Shreya Kadambi, Arash Behboodi, Joseph B. Soriaga, Max Welling, Roohollah Amiri, Srinivas Yerramalli, Taesang Yoo
2022 arXiv   pre-print
We present a neural network architecture for jointly learning user locations and environment mapping up to isometry, in an unsupervised way, from channel state information (CSI) values with no location  ...  The neural network task is set prediction and is accordingly trained end-to-end. The proposed model learns an interpretable latent, i.e., user location, by just enforcing a physics-based decoder.  ...  Many of probabilistic solutions to SLAM such as EKF-SLAM, Fast-SLAM, or Graph-SLAM have been used in multipath assisted positioning such as works in [2] , [3] , [4] .  ... 
arXiv:2203.08264v1 fatcat:zrs3ofa7v5c3tehgzo5vf6jwcy

Robust Ultra-wideband Range Error Mitigation with Deep Learning at the Edge [article]

Simone Angarano, Vittorio Mazzia, Francesco Salvetti, Giovanni Fantin, Marcello Chiaberge
2021 arXiv   pre-print
This article proposes an efficient representation learning methodology that exploits the latest advancement in deep learning and graph optimization techniques to achieve effective ranging error mitigation  ...  Indeed, multipath effects, reflections, refractions, and complexity of the indoor radio environment can easily introduce a positive bias in the ranging measurement, resulting in highly inaccurate and unsatisfactory  ...  Graph optimization (G.O.) and weight precision (W.P.) reduction further increase the capability of our already efficient neural network design helping to deal with energy, speed, size and cost constraints  ... 
arXiv:2011.14684v2 fatcat:iz5kqinydjd4rczdhlhogm2pi4

E-Commerce Picture Text Recognition Information System Based on Deep Learning

Bin Zhao, WenYing Li, Qian Guo, RongRong Song, Bai Yuan Ding
2022 Computational Intelligence and Neuroscience  
In terms of target recognition, compared with the traditional MWI-DenseNet neural network, the computation amount of the improved MWI DenseNet neural network is significantly reduced under different shunt  ...  For the accuracy requirements of commodity image detection and classification, the FPN network is improved by DPFM ablation and RFM, so as to improve the detection accuracy of commodities by the network  ...  But in the application of neural networks, increasing the network layers and amplifying the channels in the network feature graph, the effect of "widening and deepening" of convolutional neural network  ... 
doi:10.1155/2022/9474245 pmid:35106064 pmcid:PMC8801320 fatcat:wpygvyragrgt5h3kvba4hs7gfy

Deep-Waveform: A Learned OFDM Receiver Based on Deep Complex-valued Convolutional Networks [article]

Zhongyuan Zhao, Mehmet C. Vuran, Fujuan Guo, Stephen D. Scott
2021 arXiv   pre-print
In response, guidelines of exact and approximate implementations of a complex-valued convolutional layer are provided for the design and analysis of convolutional networks for wireless PHY.  ...  The proposed approach benefits from the expressive nature of complex-valued neural networks, which, however, currently lack support from popular deep learning platforms.  ...  Regarding technical solutions, convolutional neural networks (CNNs) [4] , [5] , [16] , [23] , [24] , [28] are less often used than multilayer perceptron (MLP) [6] - [12] , [21] , [22] , [25  ... 
arXiv:1810.07181v6 fatcat:nelcdghnbzhcrhrvcu2fmo5sve

MHA-Net: Multipath Hybrid Attention Network for building footprint extraction from high-resolution remote sensing imagery

Jihong Cai, Yimin Chen
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
The MHA-Net architecture consists of three components: the encoding network, multipath hybrid dilated convolution (HDC), and dense upsampling convolution (DUC).  ...  We propose a novel multipath hybrid attention network (MHA-Net) to address these challenges.  ...  In recent years, deep learning, especially the convolution neural networks (CNNs), has become one of the most prevalent methods in the computer vision field.  ... 
doi:10.1109/jstars.2021.3084805 fatcat:p3ngduovz5av3bednifsyfkne4

Road Scene Recognition of Forklift AGV Equipment Based on Deep Learning

Gang Liu, Rongxu Zhang, Yanyan Wang, Rongjun Man
2021 Processes  
neural network model for the scene recognition of forklift AGV equipment in the warehouse environment.  ...  network.  ...  part of the convolutional neural network, and the information is used to expand the feature graph when the max-unpooling operation is carried out in the decoding network.  ... 
doi:10.3390/pr9111955 fatcat:g6525w4i5vgina7yvm52ly44t4

A Hybrid Approach based on Transfer and Ensemble Learning for Improving Performances of Deep Learning Models on Small Datasets

2021 Turkish Journal of Electrical Engineering and Computer Sciences  
In 7 this study, we propose a new approach that utilizes transfer learning and ensemble methods to increase the accuracy 8 rates of convolutional neural networks for classification tasks on small data  ...  To this end, we generate different-sized 9 sub-networks by fragmenting an existing large pre-trained network then gather those networks to form an ensemble. 10 For ensemble scoring, we also suggest two  ...  ImageNet classification with deep convolutional neural networks. Advances Dieleman S, Willett K, Dambre J. Rotation-invariant convolutional neural networks for galaxy morphology predic-tion.  ... 
doi:10.3906/elk-2102-101 fatcat:xtutyp7ydzgtzl42ze2lwvl5aq
« Previous Showing results 1 — 15 out of 753 results