Filters








5,162 Hits in 9.3 sec

Parameters Compressing in Deep Learning

Shiming He, Zhuozhou Li, Yangning Tang, Zhuofan Liao, Feng Li, Se-Jung Lim
<span title="">2019</span> <i title="Computers, Materials and Continua (Tech Science Press)"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/amujz7fcqna6do727z6ev3ueo4" style="color: black;">Computers Materials &amp; Continua</a> </i> &nbsp;
With the popularity of deep learning tools in image decomposition and natural language processing, how to support and store a large number of parameters required by deep learning algorithms has become  ...  To let vector take advance of the compressing performance of matrix decomposition and tensor decomposition, we use reshaping and unfolding to let vector be the input and output of Tensor-Factorized Neural  ...  [Chien and Bao (2018) ] use the tucker decomposition to replace the affline transformation in a neural network, and builds a tensor-factorized multi-layer perceptron (MLP), named tensor-factorized neural  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.32604/cmc.2020.06130">doi:10.32604/cmc.2020.06130</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/vtz2o3enxndkbazwdwd52fzm2e">fatcat:vtz2o3enxndkbazwdwd52fzm2e</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200320153807/http://tsp.techscience.com//uploads/attached/file/20191223/20191223071224_21112.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/3f/ed/3fedde21136430d12f9ff3ef657b258945664a98.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.32604/cmc.2020.06130"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Matrix and tensor decompositions for training binary neural networks [article]

Adrian Bulat and Jean Kossaifi and Georgios Tzimiropoulos and Maja Pantic
<span title="2019-04-16">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
While prior methods for neural network binarization binarize each filter independently, we propose to instead parametrize the weight tensor of each layer using matrix or tensor decomposition.  ...  This paper is on improving the training of binary neural networks in which both activations and weights are binary.  ...  Related work In this section, we review the related work, in terms of neural network architectures (2.1), network binarization (2.2) and tensor methods (2.3).  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1904.07852v1">arXiv:1904.07852v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/z37niixs7rc3bhk35ljcteec2a">fatcat:z37niixs7rc3bhk35ljcteec2a</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200923040015/https://arxiv.org/pdf/1904.07852v1.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/c1/48/c148fc2fa831e14b6754816052bc76716d68b6c9.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1904.07852v1" 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>

Compact Neural Architecture Designs by Tensor Representations

Jiahao Su, Jingling Li, Xiaoyu Liu, Teresa Ranadive, Christopher Coley, Tai-Ching Tuan, Furong Huang
<span title="2022-03-08">2022</span> <i title="Frontiers Media SA"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/jfxjod42szdexo7gatfnybn2ca" style="color: black;">Frontiers in Artificial Intelligence</a> </i> &nbsp;
We propose a framework of tensorial neural networks (TNNs) extending existing linear layers on low-order tensors to multilinear operations on higher-order tensors.  ...  With backpropagation, we can either learn TNNs from scratch or pre-trained models using knowledge distillation.  ...  We introduce a system of generalized tensor algebra, with which we derive efficient prediction and learning in tensorial neural networks (TNNs).  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3389/frai.2022.728761">doi:10.3389/frai.2022.728761</a> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pubmed/35355829">pmid:35355829</a> <a target="_blank" rel="external noopener" href="https://pubmed.ncbi.nlm.nih.gov/PMC8959219/">pmcid:PMC8959219</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/h7ifkvrj5ze7jdj6rx3emkqj7q">fatcat:h7ifkvrj5ze7jdj6rx3emkqj7q</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220310125811/https://fjfsdata01prod.blob.core.windows.net/articles/files/728761/pubmed-zip/.versions/3/.package-entries/frai-05-728761-r2/frai-05-728761.pdf?sv=2018-03-28&amp;sr=b&amp;sig=trjMiXHHFypjanjTHII9pyplrylQrv0E%2FDAyAgJpZZQ%3D&amp;se=2022-03-10T12%3A58%3A40Z&amp;sp=r&amp;rscd=attachment%3B%20filename%2A%3DUTF-8%27%27frai-05-728761.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/f6/f5/f6f50b8d0b925801a121bfc9cebb873d06947483.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3389/frai.2022.728761"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> frontiersin.org </button> </a> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8959219" title="pubmed link"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> pubmed.gov </button> </a>

Unifying Multi-Domain Multi-Task Learning: Tensor and Neural Network Perspectives [article]

Yongxin Yang, Timothy M. Hospedales
<span title="2016-11-28">2016</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Multi-domain learning aims to benefit from simultaneously learning across several different but related domains.  ...  This generalisation has two mathematically equivalent views in multi-linear algebra and gated neural networks respectively.  ...  The prediction of task (domain) i is given asŷ Figure 3 : 3 Two-sided Neural Network for Multi-Task/Multi- Figure 4 : 4 Learning multiple domains independently, versus learning with parametrised neural  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1611.09345v1">arXiv:1611.09345v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ao45l3bjazcmxmuqyrgeqjw3am">fatcat:ao45l3bjazcmxmuqyrgeqjw3am</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200924035538/https://arxiv.org/pdf/1611.09345v1.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/71/5a/715a99769239db8778f9a5370858be6f1cf3ed18.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1611.09345v1" 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>

Tensor Decomposition via Variational Auto-Encoder [article]

Bin Liu, Zenglin Xu, Yingming Li
<span title="2016-11-03">2016</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
(via Neural Networks) whose parameters can be learned from data.  ...  The proposed model takes advantages of Neural Networks and nonparametric Bayesian models, by replacing the multi-linear product in traditional Bayesian tensor decomposition with a complex nonlinear function  ...  As an effective way to tensor analysis, tensor decomposition can analyze such a high-order tensor via its low-dimension embedding.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1611.00866v1">arXiv:1611.00866v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/d4u2yzm2oveyfal5bztjoy3yye">fatcat:d4u2yzm2oveyfal5bztjoy3yye</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200923013313/https://arxiv.org/pdf/1611.00866v1.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/12/6b/126bc681cbe0f6c86798d0014825826369a873fb.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1611.00866v1" 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>

Tensor Based Knowledge Transfer Across Skill Categories for Robot Control

Chenyang Zhao, Timothy M. Hospedales, Freek Stulp, Olivier Sigaud
<span title="">2017</span> <i title="International Joint Conferences on Artificial Intelligence Organization"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/vfwwmrihanevtjbbkti2kc3nke" style="color: black;">Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence</a> </i> &nbsp;
By factorising the weights of the neural network, we are able to extract transferrable latent skills, that enable dramatic acceleration of learning in cross-task transfer.  ...  With a suitable curriculum, this allows us to learn challenging dextrous control tasks like ball-in-cup from scratch with pure reinforcement learning.  ...  Single Task Learning The network can be trained in a supervised way via learning from demonstrationr, or by RL via policy-search.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.24963/ijcai.2017/484">doi:10.24963/ijcai.2017/484</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/ijcai/ZhaoHSS17.html">dblp:conf/ijcai/ZhaoHSS17</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/5qhizl554fhytcvvx7ay3rziti">fatcat:5qhizl554fhytcvvx7ay3rziti</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20180722143712/https://www.research.ed.ac.uk/portal/files/36019188/mainReport_CR_1.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/d5/fc/d5fc3750a00c46355a32336786468d47ca74f487.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.24963/ijcai.2017/484"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

MR-GCN: Multi-Relational Graph Convolutional Networks based on Generalized Tensor Product

Zhichao Huang, Xutao Li, Yunming Ye, Michael K. Ng
<span title="">2020</span> <i title="International Joint Conferences on Artificial Intelligence Organization"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/vfwwmrihanevtjbbkti2kc3nke" style="color: black;">Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence</a> </i> &nbsp;
In this paper, we propose the Multi-Relational Graph Convolutional Network (MR-GCN) framework by developing a novel convolution operator on multi-relational graphs.  ...  In particular, our multi-dimension convolution operator extends the graph spectral analysis into the eigen-decomposition of a Laplacian tensor.  ...  They both generate new graph structures based on meta-paths, where HAN learns the importance of fixed metapaths via an attentional graph neural network and GTN automatically learns meta-paths.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.24963/ijcai.2020/175">doi:10.24963/ijcai.2020/175</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/ijcai/HuangLYN20.html">dblp:conf/ijcai/HuangLYN20</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/6xomphi3bretbl6dkokwtdod7q">fatcat:6xomphi3bretbl6dkokwtdod7q</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201104224714/https://www.ijcai.org/Proceedings/2020/0175.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/a6/03/a603ead1c9c49e3a2780506a9c82f23b8ce9ca27.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.24963/ijcai.2020/175"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Deep Multi-task Representation Learning: A Tensor Factorisation Approach [article]

Yongxin Yang, Timothy Hospedales
<span title="2017-02-16">2017</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, we present a new deep multi-task representation learning framework that learns cross-task sharing structure at every layer in a deep network.  ...  Most contemporary multi-task learning methods assume linear models. This setting is considered shallow in the era of deep learning.  ...  CONCLUSION In this paper, we propose a novel framework for end-to-end multi-task representation learning in contemporary deep neural networks.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1605.06391v2">arXiv:1605.06391v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/n5wqvr347nenzdr2sopcavoguu">fatcat:n5wqvr347nenzdr2sopcavoguu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200828231132/https://arxiv.org/pdf/1605.06391v2.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/31/ea/31ea88f29e7f01a9801648d808f90862e066f9ea.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1605.06391v2" 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>

Tensor Networks for Dimensionality Reduction and Large-scale Optimization: Part 1 Low-Rank Tensor Decompositions

Andrzej Cichocki, Namgil Lee, Ivan Oseledets, Anh-Huy Phan, Qibin Zhao, Danilo P. Mandic
<span title="">2016</span> <i title="Now Publishers"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/ka2h7lkphrfvjlabybgqbnn2jq" style="color: black;">Foundations and Trends® in Machine Learning</a> </i> &nbsp;
Machine learning and data mining algorithms are becoming increasingly important in analyzing large volume, multi-relational and multi--modal datasets, which are often conveniently represented as multiway  ...  Keywords: Tensor networks, Function-related tensors, CP decomposition, Tucker models, tensor train (TT) decompositions, matrix product states (MPS), matrix product operators (MPO), basic tensor operations  ...  the analysis of large-scale, multi-modal and multi-relational datasets.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1561/2200000059">doi:10.1561/2200000059</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ememscddezeovamsoqrcpp33z4">fatcat:ememscddezeovamsoqrcpp33z4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20180721170159/https://www.nowpublishers.com/article/DownloadSummary/MAL-059" 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/00/33/0033fabb9f9dad4bdaa81470593601d9506d2812.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1561/2200000059"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Simple and Efficient Computational Intelligence Strategies for Effective Collaborative Decisions

Emelia Opoku Aboagye, Rajesh Kumar
<span title="2019-01-21">2019</span> <i title="MDPI AG"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/hijy7jexkvcipg3tulqv73bck4" style="color: black;">Future Internet</a> </i> &nbsp;
The evaluation results from real world datasets show that, our novel deep learning multitask tensor factorization (NeuralFil) analysis is computationally less expensive, scalable and addresses the cold-start  ...  problem through explicit multi-task approach for optimal recommendation decision making.  ...  We explore the use of deep neural networks for information filtering with multi-layer perceptron (MLP) to learn the item-user interaction in this paper.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/fi11010024">doi:10.3390/fi11010024</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/frdj5shurzdsvofeqnjqppohee">fatcat:frdj5shurzdsvofeqnjqppohee</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190501032614/https://res.mdpi.com/futureinternet/futureinternet-11-00024/article_deploy/futureinternet-11-00024.pdf?filename=&amp;attachment=1" 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/b1/70/b170faa5a81e019e5770e725266af2976cb5182b.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/fi11010024"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> mdpi.com </button> </a>

MARS: Masked Automatic Ranks Selection in Tensor Decompositions [article]

Maxim Kodryan, Dmitry Kropotov, Dmitry Vetrov
<span title="2021-06-18">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Tensor decomposition methods are known to be efficient for compressing and accelerating neural networks.  ...  During training, the procedure learns binary masks over decomposition cores that "select" the optimal tensor structure.  ...  Decomposition methods cope with redundancy via an efficient representation of neural network parameters as decomposed tensors.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.10859v2">arXiv:2006.10859v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/4utssbbxz5bmvd3wjepic2fn6i">fatcat:4utssbbxz5bmvd3wjepic2fn6i</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210623231936/https://arxiv.org/pdf/2006.10859v2.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/8f/3e8f7705973df012654f6e74144ea3982c37bf0b.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.10859v2" 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>

Depth-wise Decomposition for Accelerating Separable Convolutions in Efficient Convolutional Neural Networks [article]

Yihui He, Jianing Qian, Jianren Wang
<span title="2019-10-21">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Very deep convolutional neural networks (CNNs) have been firmly established as the primary methods for many computer vision tasks.  ...  We show our approach can be further generalized to the multi-channel and multi-layer cases, based on Generalized Singular Value Decomposition (GSVD) [59].  ...  [73] , after a deep convolutional neural network is fully decomposed, we can finetune the model for ten epochs with a small learning rate 1e −4 to obtain better accuracy (Section 4.3).  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1910.09455v1">arXiv:1910.09455v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/gxzy3feejrfnlfwms32rcx3hge">fatcat:gxzy3feejrfnlfwms32rcx3hge</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200910191453/https://arxiv.org/pdf/1910.09455v1.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/1e/69/1e691550446042283628b6a3cc52fc8adefc20e3.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1910.09455v1" 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>

Incremental multi-domain learning with network latent tensor factorization [article]

Adrian Bulat and Jean Kossaifi and Georgios Tzimiropoulos and Maja Pantic
<span title="2019-11-22">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We show that leveraging tensor structure enables better performance than simply using matrix operations. Joint tensor modelling also naturally leverages correlations across different layers.  ...  Compared with previous methods which have focused on adapting each layer separately, our approach results in more compact representations for each new task/domain.  ...  Closely Related Work In this section, we review the related work on incremental multi-domain learning and tensor methods.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1904.06345v2">arXiv:1904.06345v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/5mjv6wfxubfethf3644nzqjw6i">fatcat:5mjv6wfxubfethf3644nzqjw6i</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200903163555/https://arxiv.org/pdf/1904.06345v2.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/fb/e2/fbe217f6a91daf53293b6fec5bfe48538518e76d.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1904.06345v2" 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>

Large-Scale Spectrum Occupancy Learning via Tensor Decomposition and LSTM Networks [article]

Mohsen Joneidi, Ismail Alkhouri, Nazanin Rahnavard
<span title="2019-05-10">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Next, a time-series prediction method based on CANDECOMP/PARFAC (CP) tensor decomposition and LSTM recurrent neural networks is proposed.  ...  A new paradigm for large-scale spectrum occupancy learning based on long short-term memory (LSTM) recurrent neural networks is proposed.  ...  Tensor CP Decomposition A Tensor is a multi-dimensional array.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1905.04392v1">arXiv:1905.04392v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/yh3bt2sup5flbicfpqltvq5he4">fatcat:yh3bt2sup5flbicfpqltvq5he4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200904134437/https://arxiv.org/pdf/1905.04392v1.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/a8/d9/a8d9f9f297b5a99d843c355e73f5668f3e71dd50.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1905.04392v1" 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>

IEEE Access Special Section: Sequential Data Modeling and Its Emerging Applications

Junchi Yan, Xiaoyong Pan, Liangda Li, Changsheng Li, Peng Cui, Chao Ma
<span title="">2020</span> <i title="Institute of Electrical and Electronics Engineers (IEEE)"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/q7qi7j4ckfac7ehf3mjbso4hne" style="color: black;">IEEE Access</a> </i> &nbsp;
The method takes the paper-specific features and citation traces into account and learns the neural network model in an adversarial way.  ...  In ''Modeling similarities among multi-dimensional financial time series,'' Cheng et al. proposed a tensor-based framework for capturing the intrinsic relations among multiple factors.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/access.2020.3036537">doi:10.1109/access.2020.3036537</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/kwmmcr35wjbdrjgovio2q3doui">fatcat:kwmmcr35wjbdrjgovio2q3doui</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201208152005/https://ieeexplore.ieee.org/ielx7/6287639/8948470/09277946.pdf?tp=&amp;arnumber=9277946&amp;isnumber=8948470&amp;ref=" 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/aa/54/aa54cda2e3f521895e8a2b5e688d7c59d7c424ec.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/access.2020.3036537"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> ieee.com </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 5,162 results