Filters








12 Hits in 8.8 sec

Fast and Memory Efficient Differentially Private-SGD via JL Projections [article]

Zhiqi Bu, Sivakanth Gopi, Janardhan Kulkarni, Yin Tat Lee, Judy Hanwen Shen, Uthaipon Tantipongpipat
<span title="2021-02-05">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, we present a new framework to design differentially private optimizers called DP-SGD-JL and DP-Adam-JL.  ...  Differentially Private-SGD (DP-SGD) of Abadi et al. (2016) and its variations are the only known algorithms for private training of large scale neural networks.  ...  Acknowledgements We thank Sergey Yekhanin for his constant support and encouragement during this work. We also thank Lukas Wutschitz and Osman Ramadan for helpful discussions.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2102.03013v1">arXiv:2102.03013v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ds6fmjl4nnfnholphon5uttygq">fatcat:ds6fmjl4nnfnholphon5uttygq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210209072022/https://arxiv.org/pdf/2102.03013v1.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/83/f0/83f0f0a64f573715adc544b5a9291a88a30b3798.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2102.03013v1" 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>

On the Convergence and Calibration of Deep Learning with Differential Privacy [article]

Zhiqi Bu, Hua Wang, Qi Long
<span title="2022-02-01">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In deep learning with differential privacy (DP), the neural network achieves the privacy usually at the cost of slower convergence (and thus lower performance) than its non-private counterpart.  ...  clipping, that significantly improves the convergence, as well as preserves the same DP guarantee and computational efficiency as the existing method, which we term as local clipping.  ...  Su, Janardhan Kulkarni, Om Thakkar, and Gautam Kamath for constructive and stimulating discussions, especially around the new global-clipping algorithm.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.07830v5">arXiv:2106.07830v5</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/b7zaey545nhqdnojiqsfcx5ncq">fatcat:b7zaey545nhqdnojiqsfcx5ncq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220205114607/https://arxiv.org/pdf/2106.07830v5.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/89/61/896145846f587a0f78fb7b395498b669cc38c4a4.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.07830v5" 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>

SGD_Tucker: A Novel Stochastic Optimization Strategy for Scalable Parallel Sparse Tucker Decomposition

Hao Li, Zixuan Li, Kenli Li, Jan S. Rellermeyer, Lydia Y. Chen, Keqin Li
<span title="">2020</span> <i title="Institute of Electrical and Electronics Engineers (IEEE)"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/ll6bfs5o6bahfinh3u5z2cnyoy" style="color: black;">IEEE Transactions on Parallel and Distributed Systems</a> </i> &nbsp;
The above problems prevent deep fusion of efficient computation and big data platforms.  ...  -Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in Abstract-Sparse Tucker Decomposition (STD) algorithms learn a core tensor and a group of factor matrices to  ...  and fast convergence.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/tpds.2020.3047460">doi:10.1109/tpds.2020.3047460</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/6cvd5whchncs3lj2mphuhprela">fatcat:6cvd5whchncs3lj2mphuhprela</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210715083147/https://repository.tudelft.nl/islandora/object/uuid%3Af2da307e-bed9-49cd-a95d-dd27a3f9a937/datastream/OBJ/download" 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/29/30/29308a88b66be29ce5659b3bd58a80cfa8792fc2.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/tpds.2020.3047460"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> ieee.com </button> </a>

Matrix Sketching for Secure Collaborative Machine Learning [article]

Mengjiao Zhang, Shusen Wang
<span title="2021-07-08">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Simple defenses such as dropout and differential privacy either fail to defend the attacks or seriously hurt test accuracy.  ...  Notably, DBCL does not much increase computation and communication costs and does not hurt test accuracy at all.  ...  Acknowledgements The author thanks Michael Mahoney, Richard Peng, Peter Richtárik, and David Woodruff for their helpful suggestions.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.11201v4">arXiv:1909.11201v4</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/vojylsyggfhy3amfaqzltv73cu">fatcat:vojylsyggfhy3amfaqzltv73cu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210711203232/https://arxiv.org/pdf/1909.11201v4.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/95/14/9514a0b9c2ebc46c9a992bf1a5d17ead861ab833.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.11201v4" 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>

Noisy intermediate-scale quantum (NISQ) algorithms [article]

Kishor Bharti, Alba Cervera-Lierta, Thi Ha Kyaw, Tobias Haug, Sumner Alperin-Lea, Abhinav Anand, Matthias Degroote, Hermanni Heimonen, Jakob S. Kottmann, Tim Menke, Wai-Keong Mok, Sukin Sim (+2 others)
<span title="2021-10-06">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
A universal fault-tolerant quantum computer that can solve efficiently problems such as integer factorization and unstructured database search requires millions of qubits with low error rates and long  ...  We additionally provide a comprehensive overview of various benchmarking and software tools useful for programming and testing NISQ devices.  ...  Using differentiable quantum circuits, have also proposed an interesting approach to solving NLDE via global methods. E.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2101.08448v2">arXiv:2101.08448v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/eawbcaxikjeb7d3wqbjhutjdxy">fatcat:eawbcaxikjeb7d3wqbjhutjdxy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211009172950/https://arxiv.org/pdf/2101.08448v2.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/c0/d1/c0d1a05c8be9197208cb94588b7b3bc09c9fdac7.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2101.08448v2" 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>

Explainable AI for B5G/6G: Technical Aspects, Use Cases, and Research Challenges [article]

Shen Wang, M.Atif Qureshi, Luis Miralles-Pechuaán, Thien Huynh-The, Thippa Reddy Gadekallu, Madhusanka Liyanage
<span title="2021-12-09">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Researchers expect 6G to have higher bandwidth, coverage, reliability, energy efficiency, lower latency, and, more importantly, an integrated "human-centric" network system powered by artificial intelligence  ...  , advancing justice and strong institutions, and fostering partnership at the global level.  ...  Similar mentation of scans in healthcare, detection of crop rust to RNN, Long Short-Term Memory Networks (LSTM) is via image analysis.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.04698v1">arXiv:2112.04698v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/y7ss4opmrjbsbjm3ip2vgkkgky">fatcat:y7ss4opmrjbsbjm3ip2vgkkgky</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211211102523/https://arxiv.org/pdf/2112.04698v1.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/8b/78/8b785819345bb80b87c9013fccffa6f4e5c1dce4.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.04698v1" 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>

CARS 2020—Computer Assisted Radiology and Surgery Proceedings of the 34th International Congress and Exhibition, Munich, Germany, June 23–27, 2020

<span title="">2020</span> <i title="Springer Science and Business Media LLC"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/ohr6juw5xrcrzgnsati6eq3edq" style="color: black;">International Journal of Computer Assisted Radiology and Surgery</a> </i> &nbsp;
A hybrid (analogue and digital) CARS 2020 has therefore been envisaged to take place at the University Hospital in Munich, with a balanced combination of analogue/personal and digital presentations and  ...  Preface Analogue and Digital CARS 2020 Congress The overarching purpose of the scholarly publication and communication process of IJCARS in the context of the CARS congress could be defined as: "To enable  ...  This research is an ongoing project where we are working to achieve a fully automated and validated end-product assessment using the Welling Scale.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/s11548-020-02171-6">doi:10.1007/s11548-020-02171-6</a> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pubmed/32514840">pmid:32514840</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/lyhdb2zfpjcqbf4mmbunddwroq">fatcat:lyhdb2zfpjcqbf4mmbunddwroq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201107224331/https://link.springer.com/content/pdf/10.1007/s11548-020-02171-6.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 noreferrer" href="https://doi.org/10.1007/s11548-020-02171-6"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> springer.com </button> </a>

Ranking to Learn and Learning to Rank: On the Role of Ranking in Pattern Recognition Applications [article]

Giorgio Roffo
<span title="2017-06-01">2017</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
The last decade has seen a revolution in the theory and application of machine learning and pattern recognition.  ...  Through these advancements, variable ranking has emerged as an active and growing research area and it is now beginning to be applied to many new problems.  ...  The use of SGD in the deep neural network set- ting is motivated by the high cost of running back propagation over the full training set, since SGD can overcome this cost and still lead to fast convergence  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1706.05933v1">arXiv:1706.05933v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/oc4xtmyqkvf4njpqsojewv75qu">fatcat:oc4xtmyqkvf4njpqsojewv75qu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200930213521/https://arxiv.org/pdf/1706.05933v1.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/3c/68/3c68763caa67dee55bca76f0f71dd4530f3fd57c.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1706.05933v1" 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>

MDSSF – A Federated Architecture for Product Procurement [chapter]

Jaspreet Singh Pahwa, Pete Burnap, W. A. Gray, John Miles
<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;
In the unlikely event that the author did not send a complete manuscript and there are missing pages, th ese will be noted. Also, if material had to be removed, a note will indicate the deletion.  ...  The information described using private schemas is not shared.  ...  and procurement planning and there is always a need to increase efficiency.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/11827405_79">doi:10.1007/11827405_79</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/3qjwu5ihmjgkvikejbmuuy2xqq">fatcat:3qjwu5ihmjgkvikejbmuuy2xqq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20180722045047/http://orca.cf.ac.uk/54844/1/U585233.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 noreferrer" href="https://doi.org/10.1007/11827405_79"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> springer.com </button> </a>

Deep component analysis: algorithms and applications

George Trigeorgis, Stefanos Zafeiriou, Bjoern Schuller, Engineering And Physical Sciences Research Council, Google (Firm)
<span title="2018-05-04">2018</span>
That is to incorporate the power of neural networks with the statistical intuition and the specially crafted ideas of component analysis methods.  ...  Nonetheless, they were mostly used as black-boxes and usually required orders of magnitude more training samples than their linear counterparts to achieve similar performance.  ...  We define the loss via a soft subspace orthogonality constraint between the private and shared representation of each domain.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.25560/58861">doi:10.25560/58861</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/jmnarhrwfvhpzio6t7sd2a3gda">fatcat:jmnarhrwfvhpzio6t7sd2a3gda</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200303024926/https://spiral.imperial.ac.uk:8443/bitstream/10044/1/58861/1/Trigeorgis-G-2018-PhD-Thesis.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/01/56/015692f8e57196b91b06131258e0fa6a1f2b4de0.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.25560/58861"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Learning to Find Bugs in Programs and their Documentation

Andrew Habib
<span title="">2021</span>
This dissertation argues that learning from programs and their documentation provides an effective means to prevent and detect software bugs.  ...  Although software is pervasive, almost all programs suffer from bugs and errors.  ...  • RQ 3 : How efficient is the approach? Experimental Setup We download 5,000 Java projects from Maven.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.26083/tuprints-00017377">doi:10.26083/tuprints-00017377</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/o47olqxg5rhrzoqg4j5tazo3ni">fatcat:o47olqxg5rhrzoqg4j5tazo3ni</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210209051431/https://tuprints.ulb.tu-darmstadt.de/17377/1/Andrew_Habib_PhD_Dissertation.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/ec/b8/ecb8d8c9010b2c5eca17fd18adb8d061b3885671.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.26083/tuprints-00017377"> <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>

Horatio Nelson Lay: His Role in British Relations With China 1849-1865

Jacob J. Gerson
<span title="">1966</span>
An attempt has been made in this study to trace the background, and China career, of Horatio Nelson Lay, an Englishman, who began his public service in the British consular establishment in China and who  ...  The principal conclusions conveyed by the investigation, analysis and evaluation of the larger body of evidence fall into two categories.  ...  Firstly, Lay and Osborn were implementing the project by concentrating on forming a first class, efficient flotilla.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.25501/soas.00033870">doi:10.25501/soas.00033870</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/6yxvfhpnhng2heeuri5m4zoase">fatcat:6yxvfhpnhng2heeuri5m4zoase</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201213221638/https://eprints.soas.ac.uk/33870/1/11015581.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/b8/77/b877235e3a7851bdf03fd5f63604ea583ffc4631.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.25501/soas.00033870"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>