Filters








808 Hits in 1.9 sec

Slimmable Neural Networks [article]

Jiahui Yu, Linjie Yang, Ning Xu, Jianchao Yang, Thomas Huang
<span title="2018-12-21">2018</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We present a simple and general method to train a single neural network executable at different widths (number of channels in a layer), permitting instant and adaptive accuracy-efficiency trade-offs at runtime. Instead of training individual networks with different width configurations, we train a shared network with switchable batch normalization. At runtime, the network can adjust its width on the fly according to on-device benchmarks and resource constraints, rather than downloading and
more &raquo; ... ading different models. Our trained networks, named slimmable neural networks, achieve similar (and in many cases better) ImageNet classification accuracy than individually trained models of MobileNet v1, MobileNet v2, ShuffleNet and ResNet-50 at different widths respectively. We also demonstrate better performance of slimmable models compared with individual ones across a wide range of applications including COCO bounding-box object detection, instance segmentation and person keypoint detection without tuning hyper-parameters. Lastly we visualize and discuss the learned features of slimmable networks. Code and models are available at: https://github.com/JiahuiYu/slimmable_networks
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1812.08928v1">arXiv:1812.08928v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/wrojie66gzfwlfvizv5olagosy">fatcat:wrojie66gzfwlfvizv5olagosy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200822013141/https://arxiv.org/pdf/1812.08928v1.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/d6/0f/d60f78dfe537a30619aca0bdb0d56024d6d4cc41.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1812.08928v1" 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>

Smoothing Spline Semiparametric Density Models [article]

Jian Shi, Jiahui Yu, Anna Liu, Yuedong Wang
<span title="2019-01-10">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Density estimation plays a fundamental role in many areas of statistics and machine learning. Parametric, nonparametric and semiparametric density estimation methods have been proposed in the literature. Semiparametric density models are flexible in incorporating domain knowledge and uncertainty regarding the shape of the density function. Existing literature on semiparametric density models is scattered and lacks a systematic framework. In this paper, we consider a unified framework based on
more &raquo; ... e reproducing kernel Hilbert space for modeling, estimation, computation and theory. We propose general semiparametric density models for both a single sample and multiple samples which include many existing semiparametric density models as special cases. We develop penalized likelihood based estimation methods and computational methods under different situations. We establish joint consistency and derive convergence rates of the proposed estimators for both the finite dimensional Euclidean parameters and an infinite-dimensional functional parameter. We validate our estimation methods empirically through simulations and an application.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1901.03269v1">arXiv:1901.03269v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/p2ixq3obtbftxmg2wxwqnbc3li">fatcat:p2ixq3obtbftxmg2wxwqnbc3li</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200915030415/https://arxiv.org/pdf/1901.03269v1.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/92/89/9289126fb0585df71d8204ee8f4ced566b605151.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1901.03269v1" 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>

Smoothing Spline Semiparametric Density Models

Jiahui Yu, Jian Shi, Anna Liu, Yuedong Wang
<span title="2020-08-24">2020</span> <i title="Taylor &amp; Francis"> figshare.com </i> &nbsp;
Density estimation plays a fundamental role in many areas of statistics and machine learning. Parametric, nonparametric, and semiparametric density estimation methods have been proposed in the literature. Semiparametric density models are flexible in incorporating domain knowledge and uncertainty regarding the shape of the density function. Existing literature on semiparametric density models is scattered and lacks a systematic framework. In this article, we consider a unified framework based
more &raquo; ... reproducing kernel Hilbert space for modeling, estimation, computation, and theory. We propose general semiparametric density models for both a single sample and multiple samples which include many existing semiparametric density models as special cases. We develop penalized likelihood based estimation methods and computational methods under different situations. We establish joint consistency and derive convergence rates of the proposed estimators for both finite dimensional Euclidean parameters and an infinite-dimensional functional parameter. We validate our estimation methods empirically through simulations and an application. Supplementary materials for this article are available online.
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.6084/m9.figshare.12327953.v2">doi:10.6084/m9.figshare.12327953.v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/zy23hc3u7nbx3amlgjabomh72m">fatcat:zy23hc3u7nbx3amlgjabomh72m</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200825012713/https://s3-eu-west-1.amazonaws.com/pstorage-tf-iopjsd8797887/24416919/uasa_a_1769636_sm1872.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/10.6084/m9.figshare.12327953.v2"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> figshare.com </button> </a>

Scale-wise Convolution for Image Restoration [article]

Yuchen Fan, Jiahui Yu, Ding Liu, Thomas S. Huang
<span title="2019-12-19">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
vision and image processing has greatly benefited feature engineering (Lowe 2004) , image classification (Szegedy et al. 2015) , object detection (Cai et al. 2016; Lin et al. 2017; Fu et al. 2017; Yu  ...  Scale-wise Convolutional Networks for Image Restoration The proposed SCN is built upon widely-activated residual networks for image super-resolution Fan, Yu, and Huang 2018; Fan et al. 2019) , by adding  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1912.09028v1">arXiv:1912.09028v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/oqcamspjefdchgl45reqorn7wq">fatcat:oqcamspjefdchgl45reqorn7wq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200829025317/https://arxiv.org/pdf/1912.09028v1.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/85/3c/853c7dc712a963d0d25965868974490d7b310d76.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1912.09028v1" 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>

Self-supervised Learning with Random-projection Quantizer for Speech Recognition [article]

Chung-Cheng Chiu, James Qin, Yu Zhang, Jiahui Yu, Yonghui Wu
<span title="2022-02-03">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
STREAMING MODELS The architecture we use for the streaming experiments follows a similar design as previous work for building streaming ASRs (Yu et al., 2021) .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2202.01855v1">arXiv:2202.01855v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/bfy7qyr5tvgilgs7ksgn4pdwmq">fatcat:bfy7qyr5tvgilgs7ksgn4pdwmq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220210233525/https://arxiv.org/pdf/2202.01855v1.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/e0/05/e005dc893213a4ff0076d680dedc9d5841db6d58.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2202.01855v1" 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>

Modulated Contrast for Versatile Image Synthesis [article]

Fangneng Zhan, Jiahui Zhang, Yingchen Yu, Rongliang Wu, Shijian Lu
<span title="2022-03-17">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Perceiving the similarity between images has been a long-standing and fundamental problem underlying various visual generation tasks. Predominant approaches measure the inter-image distance by computing pointwise absolute deviations, which tends to estimate the median of instance distributions and leads to blurs and artifacts in the generated images. This paper presents MoNCE, a versatile metric that introduces image contrast to learn a calibrated metric for the perception of multifaceted
more &raquo; ... image distances. Unlike vanilla contrast which indiscriminately pushes negative samples from the anchor regardless of their similarity, we propose to re-weight the pushing force of negative samples adaptively according to their similarity to the anchor, which facilitates the contrastive learning from informative negative samples. Since multiple patch-level contrastive objectives are involved in image distance measurement, we introduce optimal transport in MoNCE to modulate the pushing force of negative samples collaboratively across multiple contrastive objectives. Extensive experiments over multiple image translation tasks show that the proposed MoNCE outperforms various prevailing metrics substantially. The code is available at https://github.com/fnzhan/MoNCE.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.09333v1">arXiv:2203.09333v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/burzwa3n5zdbdiwvw6oufaw6he">fatcat:burzwa3n5zdbdiwvw6oufaw6he</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220323110016/https://arxiv.org/pdf/2203.09333v1.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/bb/8e/bb8ed4f4abcc8f28a4efad3d127646fceccb5ea8.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.09333v1" 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>

Cross-Supervised Object Detection [article]

Zitian Chen, Zhiqiang Shen, Jiahui Yu, Erik Learned-Miller
<span title="2020-06-29">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
After learning a new object category from image-level annotations (with no object bounding boxes), humans are remarkably good at precisely localizing those objects. However, building good object localizers (i.e., detectors) currently requires expensive instance-level annotations. While some work has been done on learning detectors from weakly labeled samples (with only class labels), these detectors do poorly at localization. In this work, we show how to build better object detectors from
more &raquo; ... labeled images of new categories by leveraging knowledge learned from fully labeled base categories. We call this novel learning paradigm cross-supervised object detection. We propose a unified framework that combines a detection head trained from instance-level annotations and a recognition head learned from image-level annotations, together with a spatial correlation module that bridges the gap between detection and recognition. These contributions enable us to better detect novel objects with image-level annotations in complex multi-object scenes such as the COCO dataset.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.15056v2">arXiv:2006.15056v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/l6oduwrabnhfbafbddub5nzhri">fatcat:l6oduwrabnhfbafbddub5nzhri</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200929075356/https://arxiv.org/pdf/2006.15056v2.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/0e/da/0edadf4fdfa52d0bde5f4c8749103f16ecd6d88d.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.15056v2" 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>

Generative Adversarial Networks for Image and Video Synthesis: Algorithms and Applications [article]

Ming-Yu Liu, Xun Huang, Jiahui Yu, Ting-Chun Wang, Arun Mallya
<span title="2020-08-06">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Ming-Yu Liu, Xun Huang, Ting-Chun Wang, and Arun Mallya are with NVIDIA. Jiahui Yu is with Google. Unconditional vs. Conditional GANs.  ...  Yu et al.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2008.02793v1">arXiv:2008.02793v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/6hoqech6wfgydnlkb6dlwjqsxe">fatcat:6hoqech6wfgydnlkb6dlwjqsxe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201013170025/https://arxiv.org/pdf/2008.02793v1.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/23/77/237729237fde44eb7ab8f35aafb82c9b8a816e44.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2008.02793v1" 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>

Scale-Adaptive Vehicle Tracking Algorithm in UAV Scene

HUANG Jiahui, PENG Li, XIE Linbo
<span title="">2021</span> <i title="Journal of Computer Engineering and Applications Beijing Co., Ltd., Science Press"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/w6x32qaobvgnzp3b52qrzveu3y" style="color: black;">Jisuanji kexue yu tansuo</a> </i> &nbsp;
In order to solve the problem of model drift caused by the scale change of the target vehicle when the traditional correlation filtering algorithm tracks the vehicle in the video taken by the UAV (unmanned aerial vehicle), this paper proposes an improved scale adaptive vehicle tracking algorithm. The algorithm is based on nuclear correlation filtering. By constructing a spatial tracker that distinguishes scales, this paper uses two filters, one to locate the target vehicle??s position, the
more &raquo; ... to estimate the scale of the target vehicle, in order to quickly determine target-related information and achieve scale adaptation. In addition, in order to solve the problem of poor tracking performance caused by rapid deformation of the target vehicle, color features that are less sensitive to deformation are added to increase the robustness of the filter, and the statistical color feature method is adopted, which is not restricted by template features. The improved algorithm in this paper is tested on 28 vehicle-related video sequences in OTB and UAV data sets. The average distance accuracy is 80.8%, the average success rate is 82.7%, and the FPS reaches 58.24. Experiments show that the algorithm in this paper can improve the detection and tracking effect of vehicles in the drone scene, and can effectively solve the problems caused by the scale change and rapid deformation of the target vehicle. Compared with other nuclear-related filtering algorithms, it has better tracking accuracy and real-time performance.
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3778/j.issn.1673-9418.2005047">doi:10.3778/j.issn.1673-9418.2005047</a> <a target="_blank" rel="external noopener" href="https://doaj.org/article/5b7c6eaf142047a4ae0ef7bdd951f24f">doaj:5b7c6eaf142047a4ae0ef7bdd951f24f</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/235ifuzpmbeo7cyf7kwd44lsky">fatcat:235ifuzpmbeo7cyf7kwd44lsky</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220312094307/http://fcst.ceaj.org/CN/article/downloadArticleFile.do?attachType=PDF&amp;id=2799" 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/fd/29/fd29d967304c449e638c9205f2f9618147cf4c99.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3778/j.issn.1673-9418.2005047"> <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>

Blind Image Super-Resolution via Contrastive Representation Learning [article]

Jiahui Zhang, Shijian Lu, Fangneng Zhan, Yingchen Yu
<span title="2021-07-01">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Image super-resolution (SR) research has witnessed impressive progress thanks to the advance of convolutional neural networks (CNNs) in recent years. However, most existing SR methods are non-blind and assume that degradation has a single fixed and known distribution (e.g., bicubic) which struggle while handling degradation in real-world data that usually follows a multi-modal, spatially variant, and unknown distribution. The recent blind SR studies address this issue via degradation
more &raquo; ... but they do not generalize well to multi-source degradation and cannot handle spatially variant degradation. We design CRL-SR, a contrastive representation learning network that focuses on blind SR of images with multi-modal and spatially variant distributions. CRL-SR addresses the blind SR challenges from two perspectives. The first is contrastive decoupling encoding which introduces contrastive learning to extract resolution-invariant embedding and discard resolution-variant embedding under the guidance of a bidirectional contrastive loss. The second is contrastive feature refinement which generates lost or corrupted high-frequency details under the guidance of a conditional contrastive loss. Extensive experiments on synthetic datasets and real images show that the proposed CRL-SR can handle multi-modal and spatially variant degradation effectively under blind settings and it also outperforms state-of-the-art SR methods qualitatively and quantitatively.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2107.00708v1">arXiv:2107.00708v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/25bwmxf3krd23hmnzdxspbv2k4">fatcat:25bwmxf3krd23hmnzdxspbv2k4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210708103809/https://arxiv.org/pdf/2107.00708v1.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/87/5d/875d438ca5d17ae398997ec963b57bed832ef16d.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2107.00708v1" 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>

Multimodal Image Synthesis and Editing: A Survey [article]

Fangneng Zhan, Yingchen Yu, Rongliang Wu, Jiahui Zhang, Shijian Lu
<span title="2021-12-27">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Yu, Y. Chang, S. Lu, F. Ma, and X.  ...  , Rongliang Wu, Jiahui Zhang, Shijian Lu§ Abstract—As information exists in various modalities in real world, effective interaction and fusion among multimodal  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.13592v1">arXiv:2112.13592v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hxkfyxbtbfgltju323os3xompe">fatcat:hxkfyxbtbfgltju323os3xompe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220103093453/https://arxiv.org/pdf/2112.13592v1.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/57/18/5718dce3400e8136d735c188bbb7520695dcc7b7.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2112.13592v1" 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>

Novel Porous Aluminum Nitride Monolayer Structures A First-principles Study [article]

Yanwei Luo, Jiahui Hu, Yu Jia
<span title="2020-01-17">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
By using ab inito calculations within Density Functional Theory, we have explored the possible structures and various properties of porous AlN monolayer materials. Two kinds of porous AlN monolayer are identified, and the phonon dispersion spectrum together with the ab inito molecular dynamics simulations demonstrate that their structures are stable. We further show that these AlN porous monolayers have well-defined porous nanostructures and even higher specific surface areas, namely, which can
more &raquo; ... be comparable with graphene, and can also be maintained evenly at high temperatures. Furthermore, both porous monolayers exhibit semiconductor properties with 2.89 eV and 2.86 eV indirect band gap, respectively. In addition, the electronic structures of such porous monolayers can be modulated by strains. The band gap of porous AlN monolayer experiences an indirect-direct transition when biaxial strain is applied. A moderate 9% compression can trigger this gap transition. These results indicate that the porous AlN monolayer may potentially be used for optoelectronic applications, as well as for underlying catalysts in the future.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2001.06390v1">arXiv:2001.06390v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/qcbtf6aisrar5k5e3g7kkbcmmy">fatcat:qcbtf6aisrar5k5e3g7kkbcmmy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200321053909/https://arxiv.org/ftp/arxiv/papers/2001/2001.06390.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/2001.06390v1" 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>

Conformer: Convolution-augmented Transformer for Speech Recognition [article]

Anmol Gulati, James Qin, Chung-Cheng Chiu, Niki Parmar, Yu Zhang, Jiahui Yu, Wei Han, Shibo Wang, Zhengdong Zhang, Yonghui Wu, Ruoming Pang
<span title="2020-05-16">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Recently Transformer and Convolution neural network (CNN) based models have shown promising results in Automatic Speech Recognition (ASR), outperforming Recurrent neural networks (RNNs). Transformer models are good at capturing content-based global interactions, while CNNs exploit local features effectively. In this work, we achieve the best of both worlds by studying how to combine convolution neural networks and transformers to model both local and global dependencies of an audio sequence in
more &raquo; ... parameter-efficient way. To this regard, we propose the convolution-augmented transformer for speech recognition, named Conformer. Conformer significantly outperforms the previous Transformer and CNN based models achieving state-of-the-art accuracies. On the widely used LibriSpeech benchmark, our model achieves WER of 2.1%/4.3% without using a language model and 1.9%/3.9% with an external language model on test/testother. We also observe competitive performance of 2.7%/6.3% with a small model of only 10M parameters.
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.08100v1">arXiv:2005.08100v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/6zhg5xeqzjfb5hge7fqrmfnbme">fatcat:6zhg5xeqzjfb5hge7fqrmfnbme</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200828091344/https://arxiv.org/pdf/2005.08100v1.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/01/70/0170fc76e934ee643f869df18fb617d5357e8b4e.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.08100v1" 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>

Neural Sparse Representation for Image Restoration [article]

Yuchen Fan, Jiahui Yu, Yiqun Mei, Yulun Zhang, Yun Fu, Ding Liu, Thomas S. Huang
<span title="2020-06-08">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Inspired by the robustness and efficiency of sparse representation in sparse coding based image restoration models, we investigate the sparsity of neurons in deep networks. Our method structurally enforces sparsity constraints upon hidden neurons. The sparsity constraints are favorable for gradient-based learning algorithms and attachable to convolution layers in various networks. Sparsity in neurons enables computation saving by only operating on non-zero components without hurting accuracy.
more &raquo; ... anwhile, our method can magnify representation dimensionality and model capacity with negligible additional computation cost. Experiments show that sparse representation is crucial in deep neural networks for multiple image restoration tasks, including image super-resolution, image denoising, and image compression artifacts removal. Code is available at https://github.com/ychfan/nsr
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.04357v1">arXiv:2006.04357v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/pwvwi2jv2bf6zaoguli4stwbg4">fatcat:pwvwi2jv2bf6zaoguli4stwbg4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200929201216/https://arxiv.org/pdf/2006.04357v1.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/2a/cc/2acc6675927e8d9051259d0785cd27a7b9082b92.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.04357v1" 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>

Universally Slimmable Networks and Improved Training Techniques [article]

Jiahui Yu, Thomas Huang
<span title="2019-10-20">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Yu et al. introduced switchable batch normalization that privatizes γ, β, µ, σ 2 of BN for each subnetwork.  ...  Yu et al. [25] present the initial approach to train a single neural network executable at different widths, permitting instant and adaptive accuracyefficiency trade-offs at runtime.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1903.05134v2">arXiv:1903.05134v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/vovzmh7pyjdzta7du66ji27g6q">fatcat:vovzmh7pyjdzta7du66ji27g6q</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200830211447/https://arxiv.org/pdf/1903.05134v2.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/1c/ed/1ced02aecb07086b715edea4a177b5f77f1c9bc3.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1903.05134v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 808 results