Filters








54,353 Hits in 4.8 sec

ASCAI: Adaptive Sampling for acquiring Compact AI [article]

Mojan Javaheripi and Mohammad Samragh and Tara Javidi and Farinaz Koushanfar
<span title="2019-11-15">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
The adaptively selected samples enable ASCAI to automatically learn how to tune per-layer compression hyperparameters to optimize the accuracy/model-size trade-off.  ...  This paper introduces ASCAI, a novel adaptive sampling methodology that can learn how to effectively compress Deep Neural Networks (DNNs) for accelerated inference on resource-constrained platforms.  ...  In this paper, we propose ASCAI, an adaptive sampling methodology that automates hyperparameter selection for DNN compression.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1911.06471v1">arXiv:1911.06471v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/izch6xkscvfvvj54fbcdoaqlsi">fatcat:izch6xkscvfvvj54fbcdoaqlsi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200824214338/https://arxiv.org/pdf/1911.06471v1.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/f0/88/f088021d82db2875eef3815c92f05b4fe182ac28.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1911.06471v1" 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>

Compressive Sampling and Feature Ranking Framework for Bearing Fault Classification With Vibration Signals

Hosameldin Ahmed, Asoke K. Nandi
<span title="">2018</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;
In the second step, the search for the most important features of these compressively-sampled signals is performed using feature ranking and selection techniques.  ...  The CS-based on MMV model is the first step in this framework and provides compressively-sampled signals based on compressed sampling rates.  ...  compressively-sampled signals using feature ranking and selection techniques to rank and select fewer features from the compressively-sampled signals that can sufficiently represent characteristics of  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/access.2018.2865116">doi:10.1109/access.2018.2865116</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/5254fddgxfhzpel5jgsalkfnsm">fatcat:5254fddgxfhzpel5jgsalkfnsm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190501205955/https://bura.brunel.ac.uk/bitstream/2438/16724/3/Fulltext.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/b2/a6b2f8c0cb4636bdc9435fd70f5415ccef12fce5.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/access.2018.2865116"> <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>

Comparison of Instance Selection and Construction Methods with Various Classifiers

Marcin Blachnik, Mirosław Kordos
<span title="2020-06-05">2020</span> <i title="MDPI AG"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/smrngspzhzce7dy6ofycrfxbim" style="color: black;">Applied Sciences</a> </i> &nbsp;
These are learning vector quantization based algorithms, along with the Drop2 and Drop3 . Other methods are less efficient or provide low compression ratio.  ...  The obtained results indicate that for the most of the classifiers compressing the training set affects prediction performance and only a small group of instance selection methods can be recommended as  ...  In order to make ranking for each dataset and each classifier the results obtained for particular data filters were ranked from the best to the worst in terms of classification accuracy and compression  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/app10113933">doi:10.3390/app10113933</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/wpvaswolirgntiaixhxi554hri">fatcat:wpvaswolirgntiaixhxi554hri</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200610022520/https://res.mdpi.com/d_attachment/applsci/applsci-10-03933/article_deploy/applsci-10-03933.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/8f/19/8f19b3e7b3fa27c9bb60ef246d367d4167869892.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/app10113933"> <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>

Learning Structured Ordinal Measures for Video based Face Recognition [article]

Ran He and Tieniu Tan and Larry Davis and Zhenan Sun
<span title="2015-07-09">2015</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
This paper presents a structured ordinal measure method for video-based face recognition that simultaneously learns ordinal filters and structured ordinal features.  ...  The first part learns stable ordinal filters to project video data into a large-margin ordinal space.  ...  We introduce the term compression ratio of samples for VFR, i.e., compression ratio = the number of unique samples/ the total number of samples.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1507.02380v1">arXiv:1507.02380v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/7qlxoxyprvcpjpn2jv7e64ulsu">fatcat:7qlxoxyprvcpjpn2jv7e64ulsu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200825154122/https://arxiv.org/pdf/1507.02380v1.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/a3/67/a3677ff1b5ad66db12d81b5175b21ef2ae5f0955.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1507.02380v1" 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>

Intelligent Control of a Sensor-Actuator System via Kernelized Least-Squares Policy Iteration

Bo Liu, Sanfeng Chen, Shuai Li, Yongsheng Liang
<span title="2012-02-28">2012</span> <i title="MDPI AG"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/taedaf6aozg7vitz5dpgkojane" style="color: black;">Sensors</a> </i> &nbsp;
In this paper a new framework, called Compressive Kernelized Reinforcement Learning (CKRL), for computing near-optimal policies in sequential decision making with uncertainty is proposed via incorporating  ...  KLSPI introduce kernel trick into the LSPI framework for Reinforcement Learning, often achieving faster convergence and providing automatic feature selection via various kernel sparsification approaches  ...  Preprocessing is to directly compress the feature space by feature selection. It then learns a basis of the sample (s i , a i ) from the compressed feature space, and then use it in LSPI.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/s120302632">doi:10.3390/s120302632</a> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pubmed/22736969">pmid:22736969</a> <a target="_blank" rel="external noopener" href="https://pubmed.ncbi.nlm.nih.gov/PMC3376585/">pmcid:PMC3376585</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hbehozop5rbdhkpw2lkoaisbtq">fatcat:hbehozop5rbdhkpw2lkoaisbtq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20191207095708/http://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC3376585&amp;blobtype=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/f5/ae/f5ae20c502baf961815a5178b4a272d1ef12702b.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/s120302632"> <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> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3376585" title="pubmed link"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> pubmed.gov </button> </a>

Data Compression Measures for Meta-Learning Systems

Marcin Blachnik, Mirosław Kordos, Sławomir Golak
<span title="2018-09-26">2018</span> <i title="IEEE"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/zbznsfsymfchdlmdsmo2bglth4" style="color: black;">Proceedings of the 2018 Federated Conference on Computer Science and Information Systems</a> </i> &nbsp;
However, meta-learning systems need appropriate data descriptors for proper functioning. One of them are data compression measures which can be extracted out of the instance selection methods.  ...  When we only need to estimate the classification accuracy of the model, the compression obtained from instance selection is a good approximator, but when we need to estimate other performance measures  ...  There are algorithms that allow for stronger compression at the same accuracy level, e.g. evolutionary based instance selection [6] .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.15439/2018f87">doi:10.15439/2018f87</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/fedcsis/BlachnikKG18.html">dblp:conf/fedcsis/BlachnikKG18</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/u27dra5ftfbrzfnwms7lpu7y6e">fatcat:u27dra5ftfbrzfnwms7lpu7y6e</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190429150135/https://annals-csis.org/proceedings/2018/drp/pdf/87.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/d4/71/d471eb32207b76a21b6ad26a12554760abfa6014.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.15439/2018f87"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Automotive Rank Based ELM Using Iterative Decomposition

Archana Nagelli, VIT University, Ramesh Ragala, Badarudeen Saleena, VIT University, VIT University
<span title="2019-10-31">2019</span> <i title="The Intelligent Networks and Systems Society"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/hxhhpqr6nrfs5eblsflinrd2wa" style="color: black;">International Journal of Intelligent Engineering and Systems</a> </i> &nbsp;
In order to overcome this concern, the Automotive Rank based ELM (AR-ELM) is proposed to obtain an effective tensor decomposition for diminishing the training time.  ...  Besides, the Bayesian approach is considered in this AR-ELM to remove the redundancy from the decomposed samples of the tensor.  ...  𝑚𝑖𝑛 𝜃 Automotive rank selection based ELM The typical machine learning techniques are failed to process a high amount of data.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.22266/ijies2019.1031.29">doi:10.22266/ijies2019.1031.29</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/r7tkopjxanctrnkmg6l6tfmwhy">fatcat:r7tkopjxanctrnkmg6l6tfmwhy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220309143852/http://www.inass.org/2019/2019103129.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/2e/86/2e861265dcefae60e02764343f3000a765725fdd.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.22266/ijies2019.1031.29"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

A Selective Sampling Strategy for Label Ranking [chapter]

Massih Amini, Nicolas Usunier, François Laviolette, Alexandre Lacasse, Patrick Gallinari
<span title="">2006</span> <i title="Springer Berlin Heidelberg"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/2w3awgokqne6te4nvlofavy5a4" style="color: black;">Lecture Notes in Computer Science</a> </i> &nbsp;
We propose a novel active learning strategy based on the compression framework of [9] for label ranking functions which, given an input instance, predict a total order over a predefined set of alternatives  ...  random and heuristic-based sampling strategies.  ...  In this paper, we propose a new selective sampling strategy for label ranking.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/11871842_7">doi:10.1007/11871842_7</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/y4eiq5yuxrbd5i7xomxgcl5tmm">fatcat:y4eiq5yuxrbd5i7xomxgcl5tmm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20180728052650/https://link.springer.com/content/pdf/10.1007%2F11871842_7.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/9b/bb/9bbb7ff8d81ff94a3815e7dfacee2bb74191b739.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/11871842_7"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> springer.com </button> </a>

Efficient Heuristic Hypothesis Ranking

S. Chien, A. Stechert, D. Mutz
<span title="1999-06-01">1999</span> <i title="AI Access Foundation"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/4ax4efcwajcgvidb6hcg6mwx4a" style="color: black;">The Journal of Artificial Intelligence Research</a> </i> &nbsp;
This paper considers the problem of learning the ranking of a set of stochastic alternatives based upon incomplete information (i.e., a limited number of samples).  ...  We describe two algorithms for hypothesis ranking and their application for the probably approximately correct (PAC) and expected loss (EL) learning criteria.  ...  of favor of earlier selection^.^ Also, it is possible to divide selection error into pairwise error unequally based on estimates of hypothesis parameters in order to reduce sampling cost (for example,  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1613/jair.615">doi:10.1613/jair.615</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/mowtkmwimnhdvlrmhl4vbfi7re">fatcat:mowtkmwimnhdvlrmhl4vbfi7re</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190224131552/http://pdfs.semanticscholar.org/5634/3993762889fb2523c739159724534b11fc8b.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/56/34/56343993762889fb2523c739159724534b11fc8b.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1613/jair.615"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> Publisher / doi.org </button> </a>

Learning Discrete Representations via Constrained Clustering for Effective and Efficient Dense Retrieval [article]

Jingtao Zhan, Jiaxin Mao, Yiqun Liu, Jiafeng Guo, Min Zhang, Shaoping Ma
<span title="2021-10-12">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Besides constrained clustering, RepCONC further adopts a vector-based inverted file system (IVF) to support highly efficient vector search on CPUs.  ...  Extensive experiments on two popular ad-hoc retrieval benchmarks show that RepCONC achieves better ranking effectiveness than competitive vector quantization baselines under different compression ratio  ...  In previous works related to joint learning with PQ [4, 35, 38] , the Index Assignments are selected based on Eq. (3) .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.05789v1">arXiv:2110.05789v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/kpce2d44wrgejnportfu45gmqe">fatcat:kpce2d44wrgejnportfu45gmqe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211014220418/https://arxiv.org/pdf/2110.05789v1.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/91/42/91429255eefe48ad140ccfaf6aa1e6be11a72a53.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.05789v1" 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>

Deep Unfolding of Iteratively Reweighted ADMM for Wireless RF Sensing [article]

Udaya S.K.P. Miriya Thanthrige, Peter Jung, Aydin Sezgin
<span title="2021-12-17">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Therefore, we propose joint rank and sparsity minimization for defect detection.  ...  Further, we propose deep learning to learn the parameters of the algorithm (i.e., algorithm unfolding) to improve the accuracy and the speed of convergence of the algorithm.  ...  CONCLUSION This paper presents a deep learning-based low-rank plus sparse recovery approach for the detection of material defects based on compressive sensing.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.03686v3">arXiv:2106.03686v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/4yegrejqvnfhjiwhv7zsttamsi">fatcat:4yegrejqvnfhjiwhv7zsttamsi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211228102657/https://arxiv.org/pdf/2106.03686v3.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/30/47/3047138245e0f80c8855e58a1ad2cff5ee8c73c4.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.03686v3" 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>

Location Discriminative Vocabulary Coding for Mobile Landmark Search

Rongrong Ji, Ling-Yu Duan, Jie Chen, Hongxun Yao, Junsong Yuan, Yong Rui, Wen Gao
<span title="2011-07-27">2011</span> <i title="Springer Nature"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/hfdglwo5wbbmta6wop52fam7a4" style="color: black;">International Journal of Computer Vision</a> </i> &nbsp;
Second, we propose to learn LDVC in each region with two schemes: (1) a Ranking Sensitive PCA and (2) a Ranking Sensitive Vocabulary Boosting.  ...  Both schemes embed location cues to learn a compact descriptor, which minimizes the retrieval ranking loss by replacing the original high-dimensional signatures.  ...  More formally, we define a unified error weighting vector [w 1 , . . . , w n sample ] to measure the ranking consistency loss for n sample conjunctive rankings.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/s11263-011-0472-9">doi:10.1007/s11263-011-0472-9</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/brazvefhczga5esado6svqvaw4">fatcat:brazvefhczga5esado6svqvaw4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20170808144701/http://eeeweba.ntu.edu.sg/computervision/Research%20Papers/2012/Location%20Discriminative%20Vocabulary%20Coding%20for%20Mobile%20Landmark%20Search.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/db/d5/dbd5e6ba77be00bbf97d834d7d47bae301684a75.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/s11263-011-0472-9"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> springer.com </button> </a>

Lossless Compression for 3DCNNs Based on Tensor Train Decomposition [article]

Dingheng Wang and Guangshe Zhao and Guoqi Li and Lei Deng and Yang Wu
<span title="2019-12-08">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We give the tensorizing for 3D convolutional kernels in TT format and investigate how to select appropriate ranks for the tensor in TT format.  ...  To miniaturize 3DCNNs for the deployment in confining environments such as embedded devices, neural network compression is a promising approach.  ...  principle to select TT ranks for a sizefixed tensor based on two bases.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1912.03647v1">arXiv:1912.03647v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/7ny3h6jnfjcwtcbxlskopnmddu">fatcat:7ny3h6jnfjcwtcbxlskopnmddu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200928214536/https://arxiv.org/pdf/1912.03647v1.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/54/82/548212ed958225ed6c73d32791ff1268647efeff.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1912.03647v1" 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>

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.  ...  In this paper, we introduce MARS -- a new efficient method for the automatic selection of ranks in general tensor decompositions.  ...  Lately, Cheng et al. [2020] proposed a reinforcement learning-based rank selection scheme for tensorized neural networks which, however, also introduces extra computational requirements by separating  ... 
<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>

Towards low bit rate mobile visual search with multiple-channel coding

Rongrong Ji, Ling-Yu Duan, Jie Chen, Hongxun Yao, Yong Rui, Shih-Fu Chang, Wen Gao
<span title="">2011</span> <i title="ACM Press"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/lahlxihmo5fhzpexw7rundu24u" style="color: black;">Proceedings of the 19th ACM international conference on Multimedia - MM &#39;11</a> </i> &nbsp;
The compression function within each channel is learnt based on a novel robust PCA scheme, with specific consideration to preserve the retrieval ranking capability of the original signature.  ...  RFID tags for mobile product search), together with the visual statistics at the reference database, to learn multiple coding channels.  ...  Ranking List for Learning: Given a channel C containing n ′ photos [I1, I2, ..., I n ′ ], we randomly sample n sample photos [I ′ 1 , I ′ 2 , ..., I ′ n sample ] as queries, which generates the following  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1145/2072298.2072372">doi:10.1145/2072298.2072372</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/mm/JiDCYRCG11.html">dblp:conf/mm/JiDCYRCG11</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ir7dmmnitvg2fgdsitm376oqde">fatcat:ir7dmmnitvg2fgdsitm376oqde</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20120130185227/http://www.ee.columbia.edu/ln/dvmm/publications/11/low_long.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/62/ae/62ae6f7d1a12199715b9409c00577336724c216b.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1145/2072298.2072372"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> acm.org </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 54,353 results