Filters








31,824 Hits in 3.9 sec

Do Vision Transformers See Like Convolutional Neural Networks? [article]

Maithra Raghu, Thomas Unterthiner, Simon Kornblith, Chiyuan Zhang, Alexey Dosovitskiy
<span title="2022-03-03">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
This raises a central question: how are Vision Transformers solving these tasks? Are they acting like convolutional networks, or learning entirely different visual representations?  ...  Convolutional neural networks (CNNs) have so far been the de-facto model for visual data.  ...  Related Work Developing non-convolutional neural networks to tackle computer vision tasks, particularly Transformer neural networks [44] has been an active area of research.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2108.08810v2">arXiv:2108.08810v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/nju5i5wbbncavpit3pi2gcbsoe">fatcat:nju5i5wbbncavpit3pi2gcbsoe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220309142325/https://arxiv.org/pdf/2108.08810v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/39/b4/39b492db00faead70bc3f4fb4b0364d94398ffdb.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2108.08810v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Container: Context Aggregation Network [article]

Peng Gao, Jiasen Lu, Hongsheng Li, Roozbeh Mottaghi, Aniruddha Kembhavi
<span title="2021-10-18">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Convolutional neural networks (CNNs) are ubiquitous in computer vision, with a myriad of effective and efficient variations.  ...  spatial context in a neural network stack.  ...  However, we should be aware that powerful neural networks, particularly image classification networks can be used for harmful applications like face and gender recognition.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.01401v2">arXiv:2106.01401v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/qxj3bdes2jh4dnibkdzamtw2pq">fatcat:qxj3bdes2jh4dnibkdzamtw2pq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211020185554/https://arxiv.org/pdf/2106.01401v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/1f/66/1f668aebd03b5150c2c2fddae48f4a65cb4a80a8.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.01401v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Are Convolutional Neural Networks or Transformers more like human vision? [article]

Shikhar Tuli, Ishita Dasgupta, Erin Grant, Thomas L. Griffiths
<span title="2021-07-01">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Our focus is on comparing a suite of standard Convolutional Neural Networks (CNNs) and a recently-proposed attention-based network, the Vision Transformer (ViT), which relaxes the translation-invariance  ...  Modern machine learning models for computer vision exceed humans in accuracy on specific visual recognition tasks, notably on datasets like ImageNet.  ...  Further, we fine-tuned two models-a Transformer and a traditional Convolutional Neural Network (CNN)-on augmented datasets to find that this increases shape bias in both CNNs and Transformers.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2105.07197v2">arXiv:2105.07197v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/fp2zrwh3zrcbrdmujlagxqqk6y">fatcat:fp2zrwh3zrcbrdmujlagxqqk6y</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210713131254/https://arxiv.org/pdf/2105.07197v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c7/65/c7650fe09c2b34e43646e785e09aefe290247e52.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2105.07197v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Visual Attention Network [article]

Meng-Hao Guo, Cheng-Ze Lu, Zheng-Ning Liu, Ming-Ming Cheng, Shi-Min Hu
<span title="2022-03-08">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
While extremely simple, VAN outperforms the state-of-the-art vision transformers and convolutional neural networks with a large margin in extensive experiments, including image classification, object detection  ...  We further introduce a novel neural network based on LKA, namely Visual Attention Network (VAN).  ...  The convolutional neural networks (CNNs) [43, 42] , utilize local contextual information and translation invariance properties to greatly improve the effectiveness of neural networks.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2202.09741v3">arXiv:2202.09741v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/gqlascuu5vfwho73wy2msg4pym">fatcat:gqlascuu5vfwho73wy2msg4pym</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220310003401/https://arxiv.org/pdf/2202.09741v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/47/78/4778413e77ba6ad9b8feee7a91ced1ecae23bec0.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2202.09741v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

ResNeSt: Split-Attention Networks [article]

Hang Zhang, Chongruo Wu, Zhongyue Zhang, Yi Zhu, Haibin Lin, Zhi Zhang, Yue Sun, Tong He, Jonas Mueller, R. Manmatha, Mu Li, Alexander Smola
<span title="2020-12-30">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, we present a modularized architecture, which applies the channel-wise attention on different network branches to leverage their success in capturing cross-feature interactions and learning  ...  Previous models like SK-Net [38] introduced feature attention between two network branches, but their operation is not optimized for training efficiency and scaling to large neural networks.  ...  Since AlexNet [34] , deep convolutional neural networks [35] have dominated image classification.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.08955v2">arXiv:2004.08955v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/acjg7w26yveune2lio6cs6paf4">fatcat:acjg7w26yveune2lio6cs6paf4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200428201555/https://arxiv.org/pdf/2004.08955v1.pdf" title="fulltext PDF download [not primary version]" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <span style="color: #f43e3e;">&#10033;</span> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.08955v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Hierarchical Spatial Transformer Network [article]

Chang Shu, Xi Chen, Qiwei Xie, Hua Han
<span title="2018-01-30">2018</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Computer vision researchers have been expecting that neural networks have spatial transformation ability to eliminate the interference caused by geometric distortion for a long time.  ...  neural network.  ...  We achieve this process by a convolutional neural network with a hierarchical structure.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1801.09467v2">arXiv:1801.09467v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/qmvgx74iyrhfzdt2updxt7dxz4">fatcat:qmvgx74iyrhfzdt2updxt7dxz4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200916122239/https://arxiv.org/pdf/1801.09467v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/1a/49/1a498cafe1e191fce906410e99540efaa29c17e0.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1801.09467v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Learning Visual Odometry with a Convolutional Network

Kishore Konda, Roland Memisevic
<span title="">2015</span> <i title="SCITEPRESS - Science and and Technology Publications"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/tyl5kfigejfehnouax7r3tii24" style="color: black;">Proceedings of the 10th International Conference on Computer Vision Theory and Applications</a> </i> &nbsp;
The extracted representations are turned into information about changes in velocity and direction using a convolutional neural network.  ...  Representations of depth and motion are extracted by detecting synchrony across time and stereo channels using network layers with multiplicative interactions.  ...  Supervised learning using convolutional neural network Convolutional Neural Networks (CNNs) have been established as a powerful class of models for a variety of tasks including classification and regression  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.5220/0005299304860490">doi:10.5220/0005299304860490</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/visapp/KondaM15.html">dblp:conf/visapp/KondaM15</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/mcbrjphuuzbk7gnnzqigiplfe4">fatcat:mcbrjphuuzbk7gnnzqigiplfe4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20170922211405/http://www.iro.umontreal.ca/%7Ememisevr/pubs/VISAPP2015.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/3e/b6/3eb628c163458f5a2ba7202cec0e6128665dc30a.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.5220/0005299304860490"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Network Deconvolution [article]

Chengxi Ye, Matthew Evanusa, Hua He, Anton Mitrokhin, Tom Goldstein, James A. Yorke, Cornelia Fermüller, Yiannis Aloimonos
<span title="2020-02-25">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Convolution is a central operation in Convolutional Neural Networks (CNNs), which applies a kernel to overlapping regions shifted across the image.  ...  We apply our network deconvolution operation to 10 modern neural network models by replacing batch normalization within each.  ...  ON TRAINING NEURAL NETWORKS Training convolutional neural networks is analogous to a series of kernel estimation problem, where we have to solve for the kernels in each layer.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1905.11926v4">arXiv:1905.11926v4</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/r2xsc6f2bbfsxe5l2scenl2zze">fatcat:r2xsc6f2bbfsxe5l2scenl2zze</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200321111018/https://arxiv.org/pdf/1905.11926v4.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1905.11926v4" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Normalized Convolutional Neural Network [article]

Dongsuk Kim and Geonhee Lee and Myungjae Lee and Shin Uk Kang and Dongmin Kim
<span title="2020-05-18">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Since NC don't need other normalization layer, NCNN looks like convolutional version of Self Normalizing Network.(SNN).  ...  In this paper, we propose Normalized Convolutional Neural Network(NCNN). NCNN is more fitted to a convolutional operator than other nomralizaiton methods.  ...  to converge to better minimum point of loss function of deep neural networks.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.05274v3">arXiv:2005.05274v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/4csjltbanzcsxdy6rkh24bk7wu">fatcat:4csjltbanzcsxdy6rkh24bk7wu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200528210635/https://arxiv.org/pdf/2005.05274v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.05274v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

A Sparse Coding Interpretation of Neural Networks and Theoretical Implications [article]

Joshua Bowren
<span title="2021-08-18">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Neural networks, specifically deep convolutional neural networks, have achieved unprecedented performance in various computer vision tasks, but the rationale for the computations and structures of successful  ...  Finally we motivate potentially more robust forward transformations by maintaining sparse priors in convolutional neural networks as well performing a stronger nonlinear transformation.  ...  Introduction Advances in deep neural networks, especially deep convolutional neural networks, have made neural networks one of the premier approaches to various computer vision tasks such as This work  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2108.06622v2">arXiv:2108.06622v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/y6tsui52szhw7fyryilcu6vt54">fatcat:y6tsui52szhw7fyryilcu6vt54</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210909174808/https://arxiv.org/pdf/2108.06622v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/77/a5/77a52c3bbc86972b96d15df0b5f56ae7aba70505.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2108.06622v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Residual Networks Behave Like Ensembles of Relatively Shallow Networks [article]

Andreas Veit, Michael Wilber, Serge Belongie
<span title="2016-10-27">2016</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Further, a lesion study reveals that these paths show ensemble-like behavior in the sense that they do not strongly depend on each other.  ...  Finally, and most surprising, most paths are shorter than one might expect, and only the short paths are needed during training, as longer paths do not contribute any gradient.  ...  Acknowledgements We would like to thank Sam Kwak and Theofanis Karaletsos for insightful feedback.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1605.06431v2">arXiv:1605.06431v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/5ubirsxbyfhtriy3zcb2wvnexm">fatcat:5ubirsxbyfhtriy3zcb2wvnexm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20191027164608/https://arxiv.org/pdf/1605.06431v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/94/41/94419cbb932fa1ce9b5e89ece89ece48963dfb28.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1605.06431v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

FCNN: Fourier Convolutional Neural Networks [chapter]

Harry Pratt, Bryan Williams, Frans Coenen, Yalin Zheng
<span title="">2017</span> <i title="Springer International Publishing"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/2w3awgokqne6te4nvlofavy5a4" style="color: black;">Lecture Notes in Computer Science</a> </i> &nbsp;
Convolutional Neural Networks (CNNs) use machine learning to achieve state-of-the-art results with respect to many computer vision tasks.  ...  In this paper a Fourier Convolution Neural Network (FCNN) is proposed whereby training is conducted entirely within the Fourier domain.  ...  Acknowledgement The authors would like to acknowledge everyone in the Centre for Research in Image Analysis (CRiA) imaging team at the Institute of Ageing and Chronic Disease at the University of Liverpool  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/978-3-319-71249-9_47">doi:10.1007/978-3-319-71249-9_47</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/v3ldamlsazb3xjkztthm2ormqi">fatcat:v3ldamlsazb3xjkztthm2ormqi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190216072400/http://pdfs.semanticscholar.org/0200/506b4a0b582859ef24b9a946871d29dde0b4.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/02/00/0200506b4a0b582859ef24b9a946871d29dde0b4.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/978-3-319-71249-9_47"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> springer.com </button> </a>

Very Deep Convolutional Networks for Text Classification

Alexis Conneau, Holger Schwenk, Loïc Barrault, Yann Lecun
<span title="">2017</span> <i title="Association for Computational Linguistics"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/o3r3buwoongbbjvsatp4bckpxm" style="color: black;">Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers</a> </i> &nbsp;
The dominant approach for many NLP tasks are recurrent neural networks, in particular LSTMs, and convolutional neural networks.  ...  However, these architectures are rather shallow in comparison to the deep convolutional networks which have pushed the state-of-the-art in computer vision.  ...  Convolutional neural networks, in short Con-vNets, are very successful in computer vision.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/e17-1104">doi:10.18653/v1/e17-1104</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/eacl/SchwenkBCL17.html">dblp:conf/eacl/SchwenkBCL17</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/uawf445nezbcjazpwghlatelni">fatcat:uawf445nezbcjazpwghlatelni</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200507173221/http://eprints.whiterose.ac.uk/154414/1/conneau_eacl2017.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/79/1d/791dded78231a05c17aadef43fa11370082d8b59.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/e17-1104"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Very Deep Convolutional Networks for Text Classification [article]

Alexis Conneau, Holger Schwenk, Loïc Barrault, Yann Lecun
<span title="2017-01-27">2017</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
The dominant approach for many NLP tasks are recurrent neural networks, in particular LSTMs, and convolutional neural networks.  ...  However, these architectures are rather shallow in comparison to the deep convolutional networks which have pushed the state-of-the-art in computer vision.  ...  Convolutional neural networks, in short Con-vNets, are very successful in computer vision.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1606.01781v2">arXiv:1606.01781v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/3cr67tgtjvdt7kpc7dg2dez3tq">fatcat:3cr67tgtjvdt7kpc7dg2dez3tq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200916053741/https://arxiv.org/pdf/1606.01781v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/f7/97/f797fd44b9ddd5845611eb7a705ca9464a8819d1.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1606.01781v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Spatial Graph Convolutional Networks [article]

Tomasz Danel, Przemysław Spurek, Jacek Tabor, Marek Śmieja, Łukasz Struski, Agnieszka Słowik, Łukasz Maziarka
<span title="2020-07-02">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Our contribution is threefold: we propose a GCN-inspired architecture which (i) leverages node positions, (ii) is a proper generalization of both GCNs and Convolutional Neural Networks (CNNs), (iii) benefits  ...  Graph Convolutional Networks (GCNs) have recently become the primary choice for learning from graph-structured data, superseding hash fingerprints in representing chemical compounds.  ...  Introduction Convolutional Neural Network (CNNs) use trainable filters to process images or grid-like objects in general.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.05310v2">arXiv:1909.05310v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/r3bqgzw2qbgehe4fyeymiivkda">fatcat:r3bqgzw2qbgehe4fyeymiivkda</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200825192126/https://arxiv.org/pdf/1909.05310v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/d2/bf/d2bfab1cd36894b97c7e91bcdcdce00c9ad0c611.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.05310v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 31,824 results