Filters








89 Hits in 1.5 sec

Learning Fast Approximations of Sparse Nonlinear Regression [article]

Yuhai Song, Zhong Cao, Kailun Wu, Ziang Yan, Changshui Zhang
<span title="2020-10-26">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
The idea of unfolding iterative algorithms as deep neural networks has been widely applied in solving sparse coding problems, providing both solid theoretical analysis in convergence rate and superior empirical performance. However, for sparse nonlinear regression problems, a similar idea is rarely exploited due to the complexity of nonlinearity. In this work, we bridge this gap by introducing the Nonlinear Learned Iterative Shrinkage Thresholding Algorithm (NLISTA), which can attain a linear
more &raquo; ... nvergence under suitable conditions. Experiments on synthetic data corroborate our theoretical results and show our method outperforms state-of-the-art methods.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2010.13490v1">arXiv:2010.13490v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/cbivytoiwvc2ha7x3l2m2psi3e">fatcat:cbivytoiwvc2ha7x3l2m2psi3e</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201029064718/https://arxiv.org/pdf/2010.13490v1.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/2c/8a/2c8aa2f76b197f871c90edf6696c53841b84215d.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2010.13490v1" 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>

Weakly- and Semi-Supervised Object Detection with Expectation-Maximization Algorithm [article]

Ziang Yan, Jian Liang, Weishen Pan, Jin Li, Changshui Zhang
<span title="2017-02-28">2017</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Object detection when provided image-level labels instead of instance-level labels (i.e., bounding boxes) during training is an important problem in computer vision, since large scale image datasets with instance-level labels are extremely costly to obtain. In this paper, we address this challenging problem by developing an Expectation-Maximization (EM) based object detection method using deep convolutional neural networks (CNNs). Our method is applicable to both the weakly-supervised and
more &raquo; ... upervised settings. Extensive experiments on PASCAL VOC 2007 benchmark show that (1) in the weakly supervised setting, our method provides significant detection performance improvement over current state-of-the-art methods, (2) having access to a small number of strongly (instance-level) annotated images, our method can almost match the performace of the fully supervised Fast RCNN. We share our source code at https://github.com/ZiangYan/EM-WSD.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1702.08740v1">arXiv:1702.08740v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hpfyzk5pinhybikiynup4qqldi">fatcat:hpfyzk5pinhybikiynup4qqldi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200831105521/https://arxiv.org/pdf/1702.08740v1.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/e9/c0/e9c035af18904702696f724258070cfe7c590346.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1702.08740v1" 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>

Cloning human dental pulp cells and studying inter-clone diversity [article]

Linna Guo, Ziang Zou, Ming Yan, Marcus Freytag, Reinhard E Friedrich, Lan Kluwe
<span title="2019-07-15">2019</span> <i title="Cold Spring Harbor Laboratory"> bioRxiv </i> &nbsp; <span class="release-stage" >pre-print</span>
AbstractHeterogeneity within a putative stem cell population presents a challenge for studies and applications of such cells. Cloning may provide a strategy for reducing heterogeneity. However, previous studies have the weakness in reliability of single-cell-origin of the colonies. The present study aims to apply an alternative method to obtain clonal dental pulp cells with increased reliability of single cell origin. Dental pulp cells were cultured from 13 human wisdom teeth. Primary cultures
more &raquo; ... f 3 human benign tumors were included as comparison. Cells were seeded into wells of a 96-plate at a mean density of 1 cell/well. On the next day, wells were inspected one by one to identify wells with single cells which were followed for 3 weeks. Survived clones were expanded and further characterized. Single cells were observed in all cases, the number of single-cell-wells varied from 16 to 33. Three weeks later, survived and grown clonal cells were observed in 10 to 29 wells, giving surviving rates of 33-91%. By contrast, though single tumor cells were also observed, none of them survived. Expanded clones exhibited diversity in viability and osteogenic differentiation which also differed from their parental cells. Seeding cells at clonal density into physically separated compartments like wells in a 96 plate and comprehensive observation provides a practical strategy for increasing reliability of single-cell origin of clonal cells. Cellular heterogeneity seems to be an intrinsic feature of dental pulp cells.
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1101/703280">doi:10.1101/703280</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/2u2mg4b4cnhz5bnuncluhztdli">fatcat:2u2mg4b4cnhz5bnuncluhztdli</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200310175324/https://www.biorxiv.org/content/biorxiv/early/2019/07/15/703280.full.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/d9/7e/d97eb06739ff08d6d8135f8483615cccb5b24bab.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1101/703280"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> biorxiv.org </button> </a>

Sill-Net: Feature Augmentation with Separated Illumination Representation [article]

Haipeng Zhang, Zhong Cao, Ziang Yan, Changshui Zhang
<span title="2021-05-24">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Yan and C.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2102.03539v2">arXiv:2102.03539v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/doseqoy3irbh7kwka7nxe7sofq">fatcat:doseqoy3irbh7kwka7nxe7sofq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210602210644/https://arxiv.org/pdf/2102.03539v2.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/92/fb/92fbada297c1f1ef1b176f84e89ab177ef9cb6eb.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2102.03539v2" 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>

FMA-ETA: Estimating Travel Time Entirely Based on FFN With Attention [article]

Yiwen Sun, Yulu Wang, Kun Fu, Zheng Wang, Ziang Yan, Changshui Zhang, Jieping Ye
<span title="2020-06-07">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Estimated time of arrival (ETA) is one of the most important services in intelligent transportation systems and becomes a challenging spatial-temporal (ST) data mining task in recent years. Nowadays, deep learning based methods, specifically recurrent neural networks (RNN) based ones are adapted to model the ST patterns from massive data for ETA and become the state-of-the-art. However, RNN is suffering from slow training and inference speed, as its structure is unfriendly to parallel
more &raquo; ... To solve this problem, we propose a novel, brief and effective framework mainly based on feed-forward network (FFN) for ETA, FFN with Multi-factor self-Attention (FMA-ETA). The novel Multi-factor self-attention mechanism is proposed to deal with different category features and aggregate the information purposefully. Extensive experimental results on the real-world vehicle travel dataset show FMA-ETA is competitive with state-of-the-art methods in terms of the prediction accuracy with significantly better inference speed.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.04077v1">arXiv:2006.04077v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/b3meecb3pfhuxac4clhf6tm5j4">fatcat:b3meecb3pfhuxac4clhf6tm5j4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200610020711/https://arxiv.org/pdf/2006.04077v1.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/2006.04077v1" 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>

Prediction and Characterization of Disorder-Order Transition Regions in Proteins by Deep Learning [article]

Ziang Yan, Satoshi Omori, Kazunori D Yamada, Hafumi Nishi, Kengo Kinoshita
<span title="2021-06-11">2021</span> <i title="Cold Spring Harbor Laboratory"> bioRxiv </i> &nbsp; <span class="release-stage" >pre-print</span>
., 2014; Yan, et al., 2016) .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1101/2021.06.11.448022">doi:10.1101/2021.06.11.448022</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/aksbxs6iijfunean3qsgsc5gle">fatcat:aksbxs6iijfunean3qsgsc5gle</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210718002432/https://www.biorxiv.org/content/biorxiv/early/2021/06/11/2021.06.11.448022.full.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/15/05/150598a12db78fa5bc3cd17afc66bf65e78a249c.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1101/2021.06.11.448022"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> biorxiv.org </button> </a>

Cosmic star formation history with tomographic CIB-galaxy cross-correlation [article]

Ziang Yan, Ludovic van Waerbeke, Angus H. Wright, Maciej Bilicki, Shiming Gu, Hendrik Hildebrandt, Abhishek S. Maniyar, Tilman Tröster
<span title="2022-04-04">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Yan et al. (2021) constrained the linear galaxy bias for KiDS using galaxy-CMB lensing cross-correlations, assuming a linear model.  ...  We follow Yan et al. (2021) to correct this effect, and note that this correction has a negligible impact on our best-fit CIB parameters.  ...  Appendix A: The jackknife covariance matrix An alternative method to estimate the covariance matrix is to use jackknife resampling, as is used in Yan et al. (2021) .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2204.01649v1">arXiv:2204.01649v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/drkrdttkuzeq3efeys6cg26coy">fatcat:drkrdttkuzeq3efeys6cg26coy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220413012844/https://arxiv.org/pdf/2204.01649v1.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/30/70/30704704b02a53ef6874e3d56ddc086c3a99bdd1.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2204.01649v1" 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 Defense: Training DNNs with Improved Adversarial Robustness [article]

Ziang Yan, Yiwen Guo, Changshui Zhang
<span title="2018-12-20">2018</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Despite the efficacy on a variety of computer vision tasks, deep neural networks (DNNs) are vulnerable to adversarial attacks, limiting their applications in security-critical systems. Recent works have shown the possibility of generating imperceptibly perturbed image inputs (a.k.a., adversarial examples) to fool well-trained DNN classifiers into making arbitrary predictions. To address this problem, we propose a training recipe named "deep defense". Our core idea is to integrate an adversarial
more &raquo; ... perturbation-based regularizer into the classification objective, such that the obtained models learn to resist potential attacks, directly and precisely. The whole optimization problem is solved just like training a recursive network. Experimental results demonstrate that our method outperforms training with adversarial/Parseval regularizations by large margins on various datasets (including MNIST, CIFAR-10 and ImageNet) and different DNN architectures. Code and models for reproducing our results are available at https://github.com/ZiangYan/deepdefense.pytorch
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1803.00404v3">arXiv:1803.00404v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/kqo5qp5zkfdrhmenqitkjkljv4">fatcat:kqo5qp5zkfdrhmenqitkjkljv4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20191013221103/https://arxiv.org/pdf/1803.00404v3.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/9a/8c/9a8c5881a1f8502a79f9500865a8d2a9a06c4d42.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1803.00404v3" 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>

Adversarial Margin Maximization Networks

Ziang Yan, Yiwen Guo, Changshui Zhang
<span title="2021-03-05">2019</span> <i title="Institute of Electrical and Electronics Engineers (IEEE)"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/3px634ph3vhrtmtuip6xznraqi" style="color: black;">IEEE Transactions on Pattern Analysis and Machine Intelligence</a> </i> &nbsp;
This paper substantially extends the work of Yan et al. published at NeurIPS 2018 [11] . The original concept of the large margin principle dates back to last century.  ...  Yan Unlike the well-known scheme in linear classification [7] , the geometric margin of nonlinear DNNs scarcely has closeform solutions, making it non-trivial to get incorporated in the training objective  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/tpami.2019.2948348">doi:10.1109/tpami.2019.2948348</a> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pubmed/31634825">pmid:31634825</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/3xrmkjcc7vdqdhlehmibuzcqei">fatcat:3xrmkjcc7vdqdhlehmibuzcqei</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200906003513/https://arxiv.org/pdf/1911.05916v1.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> <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/eb/5f/eb5f246f7dd0670a9e1b03ff5d0a88ef95fc13d9.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/tpami.2019.2948348"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> ieee.com </button> </a>

An Assessment of Contamination in the Thermal-SZ Map Using Cross-correlations

Ziang Yan, Alireza Hojjati, Tilman Tröster, Gary Hinshaw, Ludovic van Waerbeke
<span title="2019-10-18">2019</span> <i title="American Astronomical Society"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/otgg2yqymve23nrsflax26msgm" style="color: black;">Astrophysical Journal</a> </i> &nbsp;
We search for potential galactic and extragalactic dust contamination in thermal Sunyaev-Zeldovich (tSZ) maps derived from the Planck data. To test for contamination, we apply a variety of galactic dust and cosmic infrared background (CIB) models to the data as part of the y map reconstruction process. We evaluate the level of contamination by cross-correlating these y maps with mass tracers based on weak lensing data. The lensing data we use are the convergence map, κ, from the Red Sequence
more &raquo; ... ster Lensing survey (RCSLens), and the CMB lensing potential map, φ, from the Planck Collaboration. We make a CIB-subtracted y map and measure the cross-correlation between it and the lensing data. By comparing it with CIB-contaminated cross-correlation, we find that the crosscorrelation between κ and y is only slightly contaminated by CIB signal, at the level of 6.8 ± 3.5 %, which implies that previous detections of κ × y are robust to CIB contamination. However, we find that φ × y is more significantly contaminated, by 16.7 ± 3.5 %, because the CMB lensing potential probes higher redshift sources that overlap more with the CIB sources. We find that Galactic dust does not significantly contaminate either cross-correlation signal.
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3847/1538-4357/ab40b2">doi:10.3847/1538-4357/ab40b2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/timcicqnr5amxl3kzpgjfgxxqu">fatcat:timcicqnr5amxl3kzpgjfgxxqu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200323141333/http://inspirehep.net/record/1695680/files/1809.09636.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/86/cd/86cd0c94600d080916bf27e4dfd95827ecbb7915.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3847/1538-4357/ab40b2"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Neural Network Architecture Optimization through Submodularity and Supermodularity [article]

Junqi Jin, Ziang Yan, Kun Fu, Nan Jiang, Changshui Zhang
<span title="2018-02-21">2018</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Deep learning models' architectures, including depth and width, are key factors influencing models' performance, such as test accuracy and computation time. This paper solves two problems: given computation time budget, choose an architecture to maximize accuracy, and given accuracy requirement, choose an architecture to minimize computation time. We convert this architecture optimization into a subset selection problem. With accuracy's submodularity and computation time's supermodularity, we
more &raquo; ... opose efficient greedy optimization algorithms. The experiments demonstrate our algorithm's ability to find more accurate models or faster models. By analyzing architecture evolution with growing time budget, we discuss relationships among accuracy, time and architecture, and give suggestions on neural network architecture design.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1609.00074v3">arXiv:1609.00074v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/wvfmodra55fwxiumwhyiwcftyu">fatcat:wvfmodra55fwxiumwhyiwcftyu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20191022111654/https://arxiv.org/pdf/1609.00074v2.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> <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/16/78/167814c6ea3195c613690c632a7019a8a3b3d4a9.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1609.00074v3" 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>

Optimizing Recurrent Neural Networks Architectures under Time Constraints [article]

Junqi Jin, Ziang Yan, Kun Fu, Nan Jiang, Changshui Zhang
<span title="2018-02-21">2018</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Recurrent neural network (RNN)'s architecture is a key factor influencing its performance. We propose algorithms to optimize hidden sizes under running time constraint. We convert the discrete optimization into a subset selection problem. By novel transformations, the objective function becomes submodular and constraint becomes supermodular. A greedy algorithm with bounds is suggested to solve the transformed problem. And we show how transformations influence the bounds. To speed up
more &raquo; ... , surrogate functions are proposed which balance exploration and exploitation. Experiments show that our algorithms can find more accurate models or faster models than manually tuned state-of-the-art and random search. We also compare popular RNN architectures using our algorithms.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1608.07892v3">arXiv:1608.07892v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/iuiihlcmhbet7i4kcwwl5xmt7u">fatcat:iuiihlcmhbet7i4kcwwl5xmt7u</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20191016135848/https://arxiv.org/pdf/1608.07892v2.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> <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/fc/42/fc4279b2943e42d21d1844c7d08cf6beddc851f9.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1608.07892v3" 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>

An Assessment of Contamination in the thermal-SZ Map Using Cross Correlations [article]

Ziang Yan, Alireza Hojjati, Tilman Tröster, Gary Hinshaw, Ludovic van Waerbeke
<span title="2018-09-25">2018</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We search for the potential contamination in the Planck thermal Sunyaev-Zeldovich (tSZ) map by calculating the cross-correlation between the tSZ signal and weak lensing by large scale structure and the Cosmic Microwave Background (CMB). The lensing data we use is the convergence map from the Red Sequence Cluster Lensing Survey (RCSLens) and the Planck CMB lensing map. We reconstruct the tSZ y map with a Needlet Internal Linear Combination method using the HFI sky maps from the Planck satellite.
more &raquo; ... We remove the CMB signal while minimizing the residual noise. The cross correlation signal from our reconstructed y map is consistent with that from the Planck team's NILC y map. The CIB and galactic dust emission are two potential sources of contamination in the reconstructed y map. We remove the CIB signal by subtracting the CIB maps reconstructed by Planck collaboration from the raw temperature maps. We find that cross-correlation between the CIB and galactic lensing contributes to (7.5±6.0)% in the Planck NILC tSZ cross galactic lensing signal within 100<ℓ<2000, which implies that previous detections of the tSZ cross galactic lensing is robust to CIB contaminations. In contrast, the Planck NILC tSZ cross CMB lensing is biased by (18.4±2.8)% in the same ℓ range. Galactic dust contamination is tested by projecting out a grey-body dust models with different dust spectral indices. Galactic dust does not affect galactic lensing cross tSZ signal significantly.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1809.09636v1">arXiv:1809.09636v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/zjualvfnzfffhcteub7m74pbjy">fatcat:zjualvfnzfffhcteub7m74pbjy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200910104728/https://arxiv.org/pdf/1809.09636v1.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/5b/2c/5b2ca0c3ef1bfac30567f829642415f818b0cc77.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1809.09636v1" 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>

Sanguinarine protects against osteoarthritis by suppressing the expression of catabolic proteases

Yan Ma, Xuewu Sun, Kangmao Huang, Shuying Shen, Xianfeng Lin, Ziang Xie, Jiying Wang, Shunwu Fan, Jianjun Ma, Xing Zhao
<span title="2017-04-11">2017</span> <i title="Impact Journals, LLC"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/yubgl6cdcrekxpjzhshpw23l3i" style="color: black;">OncoTarget</a> </i> &nbsp;
Inflammatory cytokines play critical roles in the pathogenesis of osteoarthritis. Recent studies have demonstrated that natural active substances can serve as alternative therapeutic agents for the prevention and treatment of osteoarthritis. Sanguinarine, an alkaloid isolated from the roots of Sanguinaria canadensis, is known to have anti-inflammatory properties. The aim of the present study was to investigate the therapeutic effect of Sanguinarine against osteoarthritis. Sanguinarine inhibited
more &raquo; ... interleukin-1β-induced expression of matrix metalloproteinase 1, 3, and 13, and A disintegrin and metalloproteinase with thrombospondin motifs-5 in chondrocytes, which involved the nuclear factor-κB and c-Jun N-terminal kinase signalling pathways. Furthermore, the study of interleukin-1β-induced cartilage matrix degradation in an anterior cruciate ligament transection-induced osteoarthritis model revealed that Sanguinarine ameliorated osteoarthritis by inhibiting the expression of matrix metalloproteinase 1, 3, and 13, and A disintegrin and metalloproteinase with thrombospondin motifs-5. In conclusion, we demonstrated for the first time that Sanguinarine suppressed the expression of matrix metalloproteinase 1, 3, and 13, and A disintegrin and metalloproteinase with thrombospondin motifs-5 in vitro, ex vivo, and in vivo, indicating its potential usefulness in treating osteoarthritis.
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18632/oncotarget.17036">doi:10.18632/oncotarget.17036</a> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pubmed/28968958">pmid:28968958</a> <a target="_blank" rel="external noopener" href="https://pubmed.ncbi.nlm.nih.gov/PMC5609890/">pmcid:PMC5609890</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/il2gytdxijbqxkkibcnqwcyg5m">fatcat:il2gytdxijbqxkkibcnqwcyg5m</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20180725002053/http://www.oncotarget.com/index.php?journal=oncotarget&amp;page=article&amp;op=download&amp;path%5B%5D=17036&amp;path%5B%5D=54523" 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/10.18632/oncotarget.17036"> <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> <a target="_blank" rel="external noopener" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5609890" title="pubmed link"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> pubmed.gov </button> </a>

Influence of La Doping on Magnetic and Optical Properties of Bismuth Ferrite Nanofibers

Ziang Zhang, Haiyang Liu, Yuanhua Lin, Yan Wei, Ce-Wen Nan, Xuliang Deng
<span title="">2012</span> <i title="Hindawi Limited"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/xuurhzj2hbfpzkwr44qxsocyy4" style="color: black;">Journal of Nanomaterials</a> </i> &nbsp;
The influence of La doping on the crystal structure, ferromagnetic, and optical properties of BFO nanofibers was investigated.Bi1−xLaxFeO3ultrafine nanofibers were synthesized by the electrospinning method. The surface morphology and crystal structure of the as-spun and sintered fibers were not affected by the doping. The impurity phases of the BFO crystals were weakened with the increment of La concentration. The magnetization field curves showed that the magnetization weakened under low La
more &raquo; ... ing proportion, but strengthened with the increase of the doped proportion. The magnetization curves also showed continuous strong enhancement of ferromagnetic behavior. The results of UV-vis and photoabsorption testing revealed little influence of La doping on the optical property.
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1155/2012/238605">doi:10.1155/2012/238605</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/pvv5wq75wfhbherdidi2r5whiu">fatcat:pvv5wq75wfhbherdidi2r5whiu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190504155409/http://downloads.hindawi.com/journals/jnm/2012/238605.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/f8/79f87809a33ea718a0fb9a2b321d4857936fe13e.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1155/2012/238605"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> hindawi.com </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 89 results