Filters








3,633 Hits in 4.1 sec

Stochastic AUC Maximization with Deep Neural Networks [article]

Mingrui Liu, Zhuoning Yuan, Yiming Ying, Tianbao Yang
<span title="2020-06-30">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, we consider stochastic AUC maximization problem with a deep neural network as the predictive model.  ...  The main contribution made in this paper is to make stochastic AUC maximization more practical for deep neural networks and big data with theoretical insights as well.  ...  CONCLUSION In this paper, we consider stochastic AUC maximization problem when the predictive model is a deep neural network.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1908.10831v5">arXiv:1908.10831v5</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/qlo4tmdcwjbkjbqftddpxfofw4">fatcat:qlo4tmdcwjbkjbqftddpxfofw4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200706165710/https://arxiv.org/pdf/1908.10831v5.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/aa/c6/aac61e97a901a162ec00f18671a4152d70dc5db0.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1908.10831v5" 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>

Communication-Efficient Distributed Stochastic AUC Maximization with Deep Neural Networks [article]

Zhishuai Guo, Mingrui Liu, Zhuoning Yuan, Li Shen, Wei Liu, Tianbao Yang
<span title="2020-10-08">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, we study distributed algorithms for large-scale AUC maximization with a deep neural network as a predictive model.  ...  Although distributed learning techniques have been investigated extensively in deep learning, they are not directly applicable to stochastic AUC maximization with deep neural networks due to its striking  ...  Stochastic auc maximization with deep neural networks. ICLR, 2020b.Natole, M., Ying, Y., and Lyu, S. Stochastic proximal algorithms for auc maximization.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.02426v2">arXiv:2005.02426v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/7maye6r27jhfnptts5f3yskcdu">fatcat:7maye6r27jhfnptts5f3yskcdu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201018141145/https://arxiv.org/pdf/2005.02426v2.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/2005.02426v2" 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 AUC Maximization for Medical Image Classification: Challenges and Opportunities [article]

Tianbao Yang
<span title="2021-11-01">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Since AUC (aka area under ROC curve) is a standard performance measure for medical image classification, hence directly optimizing AUC could achieve a better performance for learning a deep neural network  ...  In this extended abstract, we will present and discuss opportunities and challenges brought about by a new deep learning method by AUC maximization (aka Deep AUC Maximization or DAM) for medical image  ...  Let h w (•) : R d → R denote a predictive model (e.g., a deep neural network, a linear model).  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2111.02400v1">arXiv:2111.02400v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/4thbkrbbqndqbk3ksaiwc3zhzu">fatcat:4thbkrbbqndqbk3ksaiwc3zhzu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211116203105/https://arxiv.org/pdf/2111.02400v1.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/80/d6/80d69029305d4a73b0241d0f9d76151e0143ab0a.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2111.02400v1" 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 and Visualizing Deep Learning for Benign/Malignant Classification in Breast Tumors [article]

Darvin Yi, Rebecca Lynn Sawyer, David Cohn III, Jared Dunnmon, Carson Lam, Xuerong Xiao, Daniel Rubin
<span title="2017-05-17">2017</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Using convolutional neural networks (CNNs), we are able to achieve an accuracy of 85% and an ROC AUC of 0.91, while leading hand-crafted feature based methods are only able to achieve an accuracy of 71%  ...  In addition, we have created a novel method to visualize what features the neural network detects for the benign/malignant classification, and have correlated those features with well known radiological  ...  Importantly, [21] report an AUC of 0.925 on the slide classification task while a pathologist reported an AUC of 0.966however, with the aid of the model's predictions, the pathologist AUC was increased  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1705.06362v1">arXiv:1705.06362v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hgtbcizzcfhltoscip6asy5xoe">fatcat:hgtbcizzcfhltoscip6asy5xoe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200929024039/https://arxiv.org/pdf/1705.06362v1.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/45/96/4596eea57d200d92eb64ef999463c7b6ea309490.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1705.06362v1" 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>

AUC Maximization in the Era of Big Data and AI: A Survey [article]

Tianbao Yang, Yiming Ying
<span title="2022-04-02">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Recently, stochastic AUC maximization for big data and deep AUC maximization for deep learning have received increasing attention and yielded dramatic impact for solving real-world problems.  ...  We also identify and discuss remaining and emerging issues for deep AUC maximization, and provide suggestions on topics for future work.  ...  ] , neural networks [157] , deep neural nets [171] .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.15046v2">arXiv:2203.15046v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/vj5gaihdc5d5lanbwehg2yh7pi">fatcat:vj5gaihdc5d5lanbwehg2yh7pi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220614155751/https://arxiv.org/pdf/2203.15046v2.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/6c/46/6c46f0b6e6299be4fd3ff6168a77d5df995f5143.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.15046v2" 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>

Supervised Anomaly Detection based on Deep Autoregressive Density Estimators [article]

Tomoharu Iwata, Yuki Yamanaka
<span title="2019-04-12">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
The proposed method effectively utilizes the anomaly label information by training the neural density estimator so that the likelihood of normal instances is maximized and the likelihood of anomalous instances  ...  By the recent advance of deep learning, the density estimation performance has been greatly improved.  ...  We optimized the neural network parameters using ADAM with learning rate 10 −3 . Table 2 shows the AUC results. The proposed method achieved the highest average AUC among the ten methods.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1904.06034v1">arXiv:1904.06034v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/uuajsm367jbpfbykbg63uftqpy">fatcat:uuajsm367jbpfbykbg63uftqpy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200930115720/https://arxiv.org/pdf/1904.06034v1.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/39/e7/39e7e879f1f2eae88084faebf64c4a9376909524.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1904.06034v1" 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>

Training distributed deep recurrent neural networks with mixed precision on GPU clusters

Alexey Svyatkovskiy, Julian Kates-Harbeck, William Tang
<span title="">2017</span> <i title="ACM Press"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/zigbcra6rjdivda6lkzknwuo5q" style="color: black;">Proceedings of the Machine Learning on HPC Environments - MLHPC&#39;17</a> </i> &nbsp;
In this paper, we evaluate training of deep recurrent neural networks with half-precision floats.  ...  We introduce a learning rate schedule facilitating neural network convergence at up to O(100) workers.  ...  of deep recurrent neural networks with half-precision floats has been evaluated on a computing framework integrating TensorFlow with custom parameter averaging and global weight update routines implemented  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1145/3146347.3146358">doi:10.1145/3146347.3146358</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/sc/SvyatkovskiyKT17.html">dblp:conf/sc/SvyatkovskiyKT17</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/3f73fxh6s5dcpdrxpbqbyan4vy">fatcat:3f73fxh6s5dcpdrxpbqbyan4vy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200830060045/https://arxiv.org/pdf/1912.00286v1.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/f7/ae/f7ae03baa092a4ba1641a8b825f16689fc5d0738.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1145/3146347.3146358"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> acm.org </button> </a>

Churn analysis using deep convolutional neural networks and autoencoders [article]

Artit Wangperawong, Cyrille Brun, Olav Laudy, Rujikorn Pavasuthipaisit
<span title="2016-04-18">2016</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Supervised learning was performed on labeled data of over 6 million customers using deep convolutional neural networks, which achieved an AUC of 0.743 on the test dataset using no more than 12 temporal  ...  Images that maximally activate the hidden units of an autoencoder trained with churned customers reveal ample opportunities for action to be taken to prevent churn among strong data, no voice users.  ...  Another strategy to improve the AUC is to pre-train the weights of the deep convolutional neural network using stacked convolutional autoencoders [13] .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1604.05377v1">arXiv:1604.05377v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/amwpazwpsfdk3czh2njxcft244">fatcat:amwpazwpsfdk3czh2njxcft244</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200913234801/https://arxiv.org/ftp/arxiv/papers/1604/1604.05377.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/04/b5/04b5ed7e8f45017806d53a2ee47d720d132b9648.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1604.05377v1" 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>

Specifying Weight Priors in Bayesian Deep Neural Networks with Empirical Bayes

Ranganath Krishnan, Mahesh Subedar, Omesh Tickoo
<span title="2020-04-03">2020</span> <i title="Association for the Advancement of Artificial Intelligence (AAAI)"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/wtjcymhabjantmdtuptkk62mlq" style="color: black;">PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE</a> </i> &nbsp;
Stochastic variational inference for Bayesian deep neural network (DNN) requires specifying priors and approximate posterior distributions over neural network weights.  ...  We propose MOdel Priors with Empirical Bayes using DNN (MOPED) method to choose informed weight priors in Bayesian neural networks.  ...  Background Bayesian neural networks Bayesian neural networks provide a probabilistic interpretation of deep learning models by placing distributions over the neural network weights (Neal 1995) .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1609/aaai.v34i04.5875">doi:10.1609/aaai.v34i04.5875</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/dfwkctj3wbcejgzy2c4b5zujva">fatcat:dfwkctj3wbcejgzy2c4b5zujva</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201103201549/https://aaai.org/ojs/index.php/AAAI/article/download/5875/5731" 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/d5/9ad52172b81b70da036ff2a941d65ea9b4f93ac7.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1609/aaai.v34i04.5875"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Advanced Graph and Sequence Neural Networks for Molecular Property Prediction and Drug Discovery [article]

Zhengyang Wang, Meng Liu, Youzhi Luo, Zhao Xu, Yaochen Xie, Limei Wang, Lei Cai, Qi Qi, Zhuoning Yuan, Tianbao Yang, Shuiwang Ji
<span title="2021-07-06">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
With the advances of deep learning methods, computational approaches for predicting molecular properties are gaining increasing momentum.  ...  In particular, our methods achieve #1 ranking in terms of both ROC-AUC and PRC-AUC on the AI Cures Open Challenge for drug discovery related to COVID-19.  ...  We also employ advanced stochastic optimization methods for optimizing ROC-AUC and PRC-AUC measures for learning deep neural networks.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.01981v3">arXiv:2012.01981v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/oldovpd55rbevbjtwdahueb4mm">fatcat:oldovpd55rbevbjtwdahueb4mm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210713225142/https://arxiv.org/pdf/2012.01981v3.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/c6/60/c66014aaaf0d38f8186edd3566fdd594b046030c.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.01981v3" 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>

BAR: Bayesian Activity Recognition using variational inference [article]

Ranganath Krishnan, Mahesh Subedar, Omesh Tickoo
<span title="2018-12-01">2018</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Applications such as video surveillance for identifying suspicious activities are designed with deep neural networks (DNNs), but DNNs do not provide uncertainty estimates.  ...  Uncertainty estimation in deep neural networks is essential for designing reliable and robust AI systems.  ...  Background Bayesian neural networks (BNNs) offer a probabilistic interpretation of deep learning models by placing distributions over the model parameters.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1811.03305v2">arXiv:1811.03305v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/3ou4opmadjcvrb6ol5njbw7j4e">fatcat:3ou4opmadjcvrb6ol5njbw7j4e</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200904020623/https://arxiv.org/pdf/1811.03305v2.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/69/87/6987aa3d6896c5837124dbb14af529133984104a.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1811.03305v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Large-scale Robust Deep AUC Maximization: A New Surrogate Loss and Empirical Studies on Medical Image Classification [article]

Zhuoning Yuan, Yan Yan, Milan Sonka, Tianbao Yang
<span title="2021-09-07">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Deep AUC Maximization (DAM) is a new paradigm for learning a deep neural network by maximizing the AUC score of the model on a dataset.  ...  Most previous works of AUC maximization focus on the perspective of optimization by designing efficient stochastic algorithms, and studies on generalization performance of large-scale DAM on difficult  ...  References [1] Deep auc maximization code. https://github.com/Optimization-AI/ICCV2021_ DeepAUC.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.03173v2">arXiv:2012.03173v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/jvbgarjcbrdy7nhxx5aadljvre">fatcat:jvbgarjcbrdy7nhxx5aadljvre</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210914190201/https://arxiv.org/pdf/2012.03173v2.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/06/5d/065d77e46d3f7f06527b504cafdc44130cfb730a.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.03173v2" 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>

Specifying Weight Priors in Bayesian Deep Neural Networks with Empirical Bayes [article]

Ranganath Krishnan and Mahesh Subedar and Omesh Tickoo
<span title="2019-12-28">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Stochastic variational inference for Bayesian deep neural network (DNN) requires specifying priors and approximate posterior distributions over neural network weights.  ...  We propose MOdel Priors with Empirical Bayes using DNN (MOPED) method to choose informed weight priors in Bayesian neural networks.  ...  Background Bayesian neural networks Bayesian neural networks provide a probabilistic interpretation of deep learning models by placing distributions over the neural network weights (Neal 1995) .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1906.05323v3">arXiv:1906.05323v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/6yipmzqaivc2vo5hdtxlye45rq">fatcat:6yipmzqaivc2vo5hdtxlye45rq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200321003532/https://arxiv.org/pdf/1906.05323v3.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/1906.05323v3" 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>

H-VGRAE: A Hierarchical Stochastic Spatial-Temporal Embedding Method for Robust Anomaly Detection in Dynamic Networks [article]

Chenming Yang, Liang Zhou, Hui Wen, Zhiheng Zhou, Yue Wu
<span title="2020-07-14">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, a stochastic neural network, named by Hierarchical Variational Graph Recurrent Autoencoder (H-VGRAE), is proposed to detect anomalies in dynamic networks by the learned robust node representations  ...  ; 2) H-VGRAE can be extended to deep structure with the increase of the dynamic network scale; 3) the anomalous edge and node can be located and interpreted from the probabilistic perspective.  ...  The encoder is implemented with Graph Neural Network (GNN), which is able to fuse the content and structural information of a node with its neighbors.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2007.06903v1">arXiv:2007.06903v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/wmjzqevpqfhfrckhuabqgqv6ey">fatcat:wmjzqevpqfhfrckhuabqgqv6ey</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200723005932/https://arxiv.org/pdf/2007.06903v1.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/2007.06903v1" 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>

Class-dependent Compression of Deep Neural Networks [article]

Rahim Entezari, Olga Saukh
<span title="2020-04-19">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Today's deep neural networks require substantial computation resources for their training, storage, and inference, which limits their effective use on resource-constrained devices.  ...  Many recent research activities explore different options for compressing and optimizing deep models.  ...  Our method automatically shrinks a trained deep neural network for mobile devices integration. In this paper, we focus on keeping the number of FN low when compressing a deep network.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.10364v3">arXiv:1909.10364v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ycmwclc5uvaf3f54j24wofk7qe">fatcat:ycmwclc5uvaf3f54j24wofk7qe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200907093538/https://arxiv.org/pdf/1909.10364v3.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/06/f5/06f56fc55b08665c88a77c92cbce40d78bdb2ed1.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.10364v3" 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 3,633 results