检索结果-内蒙古大学图书馆

27th ACM International Conference on Multimedia (MM)

作者： Huang, Tianxin Liu, Yong Zhejiang Univ State Key Lab Ind Control Technol Hangzhou Peoples R China NetEase Fuxi AI Lab Beijing Peoples R China

ISBN: (纸本)9781450368896

3D point cloud presentation has been widely used in computer vision, automatic driving, augmented reality, smart cities and virtual reality. 3D point cloud compression method with higher compression ratio and tiny loss is the key to improve data transportation efficiency. In this paper, we propose a new 3D point cloud geometry compression method based on deep learning, also an auto-encoder performing better than other networks in detail reconstruction. It can reach much higher compression ratio than the state-of-art while keeping tolerable loss. It also supports parallel compressing multiple models by GPU, which can improve processing efficiency greatly. The compression process is composed of two parts. Firstly, Raw data is compressed into codeword by extracting feature of raw model with encoder. Then, the codeword is further compressed with sparse coding. Decompression process is implemented in reverse order. Codeword is recovered and fed into decoder to reconstruct point cloud. Detail reconstruction ability is improved by a hierarchical structure in our decoder. Latter outputs are grown from former fuzzier outputs. In this way, details are added to former output by latter layers step by step to make a more precise prediction. We compare our method with PCL compression and Draco compression on ShapeNet40 part dataset. Our method may be the first deep learning-based point cloud compression algorithm. The experiments demonstrate it is superior to former common compression algorithms with large compression ratio, which can also reserve original shapes with tiny loss.

关键词： 3D point cloud geometry compression auto-encoder hierarchical structure detail reconstruction

来源：评论

学校读者我要写书评

暂无评论

Credit Risk Prediction in Peer-to-Peer Lending with Ensemble Learning Framework 31

Credit Risk Prediction in Peer-to-Peer Lending with Ensemble...

引用

31st Chinese Control And Decision Conference (CCDC)

作者： Chen, Shuhui Wang, Qing Liu, Shuan Northeastern Univ Coll Informat Sci & Engn Shenyang 110819 Peoples R China

ISBN: (纸本)9781728101057

Online Peer-to-Peer (P2P) has become a popular way of lending in recent years. Individuals can borrow or lend money directly through an online P2P platform without the help of institutional intermediaries such as banks. It is very important for the platform to predict the credit risk on whether a potential borrower will repay the loan or not, such that the defaulting of borrowers can be avoided as much as possible to keep the platform running healthily. In this paper the machine learning is adopted to design a prediction process, to solve the credit evaluation problem of online P2P lending. A multi-stage ensemble learning model is proposed to evaluate the borrowers' credits, in which the Gradient Boosting Decision Tree (GBDT) algorithm is used for feature mapping and a special auto-encoder is applied to extract the best features. Furthermore, the logical regression algorithm is designed to classify the borrowers. Through computational experiments on the open data set and taking AUC (Area Under Curve) as evaluation index, the effectiveness of the proposed model and algorithms is testified.

关键词： peer-to-peer lending ensemble learning auto-encoder logistic regression

来源：评论

学校读者我要写书评

暂无评论

Overcoming Data Limitation in Medical Visual Question Answering 22nd

Overcoming Data Limitation in Medical Visual Question Answer...

引用

10th International Workshop on Machine Learning in Medical Imaging (MLMI) / 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)

作者： Nguyen, Binh D. Thanh-Toan Do Nguyen, Binh X. Do, Tuong Tjiputra, Erman Tran, Quang D. AIOZ Pte Ltd Singapore Singapore Univ Liverpool Liverpool Merseyside England

ISBN: (纸本)9783030322519;9783030322502

Traditional approaches for Visual Question Answering (VQA) require large amount of labeled data for training. Unfortunately, such large scale data is usually not available for medical domain. In this paper, we propose a novel medical VQA framework that overcomes the labeled data limitation. The proposed framework explores the use of the unsupervised Denoising auto-encoder (DAE) and the supervised Meta-Learning. The advantage of DAE is to leverage the large amount of unlabeled images while the advantage of Meta-Learning is to learn meta-weights that quickly adapt to VQA problem with limited labeled data. By leveraging the advantages of these techniques, it allows the proposed framework to be efficiently trained using a small labeled training set. The experimental results show that our proposed method significantly outperforms the state-of-the-art medical VQA. The source code is available at https://***/aioz-ai/MICCAI19-MedVQA.

关键词： Visual Question Answering auto-encoder Meta-Learning

来源：评论

学校读者我要写书评

暂无评论

Spatial-Temporal Features Based Sensor Network Partition in Dam Safety Monitoring System

引用

SENSORS 2020年第9期20卷 2517-2517页

作者： Chen, Hao Mao, Yingchi Wang, Longbao Qi, Hai Hohai Univ Coll Water Conservancy & Hydropower Engn Nanjing 211100 Peoples R China Huaneng Lancang River Hydropower Co Ltd Ctr Res & Dev Kunming 650214 Peoples R China Hohai Univ Sch Comp & Informat Nanjing 211100 Peoples R China

Many various types of sensors have been installed to monitor the deformation and stress in the dam structure. It is difficult to directly evaluate the operation status of the dam structure based on the massive monitoring data. The sensor network is divided into multiple regions according to the design specifications, simulation data, and engineering experiences. The local results from sub-regions are integrated to achieve overall evaluation. However, it ignores the spatial distribution of sensors and the variation of time series, which cannot meet the real-time evaluation for the dam safety monitoring. If the network partitions can provide the preliminary foundation for analyzing the dynamic change laws of the dam's working conditions in a real-way, we should consider the similarity of structure and stresses in the local region of the dam and the correlation among the monitoring data. A time-series denoising autoencoder (TSDA) is proposed to represent the spatial and temporal features of the nodes by compressing high-dimensional monitoring data. Then, a network partitioning algorithm (NPA) based on spatial-temporal features based on the TSDA is presented. The NPA ensures that the partition results can support the analysis of the physical change laws by introducing the auxiliary objective variable to optimize the network partition objective function. Experimental results on the public datasets and a real dataset from an arch dam demonstrate that the proposed network partition algorithm NPA can achieve better partition performance than TSDA+K-Means and TSDA+GMM. The NPA can improve the silhouette coefficient by 45.1% and 58.4% higher than the TSDA+K-Means and TSDA+GMM, respectively. The NPA can increase the Calinski-Harabaz Index by 30.8% and 61.6%, respectively.

关键词： network partition spatial-temporal feature auto-encoder dam safety monitoring sensor networks

来源：评论

学校读者我要写书评

暂无评论

Ranking-Based Attacks to In-Region Location Verification Systems

Ranking-Based Attacks to In-Region Location Verification Sys...

引用

IEEE International Workshop on Information Forensics and Security (WIFS)

作者： Brighente, A. Formaggio, F. Ruvoletto, G. Tomasin, S. Univ Padua Dept Informat Engn Padua Italy

ISBN: (纸本)9781728132174

In-region location verification (IRLV) aims at securely verifying that a wireless device is in a given region of interest (ROI), e.g., an authorized area where specific services can be provided. We consider an IRLV system that verifies the position through the estimated attenuations of the wireless channels connecting the device to the access points (APs) of the IRLV network. We propose techniques that can be used by an attacker located outside the ROI in order to break the IRLV system and let the network believe that he is inside the ROI. This is achieved either by transmitting from positions outside the ROI exhibiting wireless channel characteristics similar to inner positions (passive attack), or by directly inducing desired estimated channel attenuations at the APs, i.e., suitably transmitting signals with multiple antennas (active attacks). In both cases, we aim at minimizing the number of attack attempts before the first successful attack. To this end we consider ranking solutions based on machine learning (ML), where the attacker ranks random candidate attacks according to metrics related their success, and then performs attacks according to the ranking.

关键词： auto-encoder in-region location verification machine learning ranking

来源：评论

学校读者我要写书评

暂无评论

ADVERSARIAL REPRESENTATION LEARNING FOR DYNAMIC SCENE DEBLURRING: A SIMPLE, FAST AND ROBUST APPROACH 26

ADVERSARIAL REPRESENTATION LEARNING FOR DYNAMIC SCENE DEBLUR...

引用

26th IEEE International Conference on Image Processing (ICIP)

作者： Liu, Yuan-Yuan Ye, Lu-Yue Shao, Wen-Ze Ge, Qi Wang, Li-Qian Bao, Bing-Kun Li, Hai-Bo Nanjing Univ Posts & Telecommun Coll Telecommun & Informat Engn 66 Xinmofan Rd Nanjing 210003 Peoples R China

ISBN: (纸本)9781538662496

In this paper, we investigate a novel learning-based method for dynamic scene deblurring. Since the inference model is formulated as an encoder-decoder, the core task has turned to learning blur-invariant hidden features from the blurred images to a great degree. To achieve state-of-the-art results in terms of both deblurring accuracy and efficiency, a simple, robust and computationally efficient deep auto-encoder is developed tailored specifically for blind deblurring, which is learned in an adversarial fashion based on use of the recent Wasserstein generative adversarial networks. Thanks to the designed framework, the new model has shown comparable or superior performance both qualitatively and quantitatively to existing state-of-the-art methods. What is more important, instead of exploiting the multi-scale strategy as previous methods, our model is just single-scale capable of achieving 6 times more efficiency than the closest competitor by Nah et al. [1], which is a more complex multi-scale deep method.

关键词： Dynamic scence deblurring adversarial training auto-encoder blind deconvolution

来源：评论

学校读者我要写书评

暂无评论

Robust Embedded Deep K-means Clustering 19

Robust Embedded Deep K-means Clustering

引用

28th ACM International Conference on Information and Knowledge Management (CIKM)

作者： Zhang, Rui Tong, Hanghang Xia, Yinglong Zhu, Yada Arizona State Univ Tempe AZ USA Univ Illinois Urbana IL 61801 USA Facebook Menlo Pk CA USA IBM TJ Watson Res Yorktown Hts NY USA

ISBN: (纸本)9781450369763

Deep neural network clustering is superior to the conventional clustering methods due to deep feature extraction and nonlinear dimensionality reduction. Nevertheless, deep neural network leads to a rough representation regarding the inherent relationship of the data points. Therefore, it is still difficult for deep neural network to exploit the effective structure for direct clustering. To address this issue, we propose a robust embedded deep K-means clustering (RED-KC) method. The proposed RED-KC approach utilizes the delta-norm metric to constrain the feature mapping process of the auto-encoder network, so that data are mapped to a latent feature space, which is more conducive to the robust clustering. Compared to the existing auto-encoder networks with the fixed prior, the proposed RED-KC is adaptive during the process of feature mapping. More importantly, the proposed RED-KC embeds the clustering process with the auto-encoder network, such that deep feature extraction and clustering can be performed simultaneously. Accordingly, a direct and efficient clustering could be obtained within only one step to avoid the inconvenience of multiple separate stages, namely, losing pivotal information and correlation. Consequently, extensive experiments are provided to validate the effectiveness of the proposed approach.

关键词： embedded clustering deep neural networks robust k-means auto-encoder

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Development Review

Deep Learning Development Review

引用

Advanced Electronic Materials, Computers and Software Engineering (AEMCSE), International Conference on

作者： Wen-hao Lv Ju-yang Lei Shanghai University of Engineering Science Shanghai China

ISBN: (数字)9781728181431

ISBN: (纸本)9781728181448

As a new branch of the machine learning, the nature of deep learning is to establish and simulate the neural network of human brain to analysis and learning. With the development of neural networks, the models are getting bigger and more complex, the network model is no longer a few layers, dozens or even hundreds of network models play a huge advantage. In recent years, various deep neural network models have achieved remarkable results in many fields, such as face recognition, voice recognition, natural language processing and so on. People called these large-scale neural networks 'deep learning'. This paper mainly reviews the development history of deep learning, and make a brief summary of the problems faced by deep learning at the end of the paper.

关键词： deep learning machine learning convolutional neural network recursive neural network auto-encoder Machine Learning network model Artificial Intelligence Neural network wetware Speech recognition Neural Networks (Computer)

来源：评论

学校读者我要写书评

暂无评论

LOW DOSE ABDOMINAL CT IMAGE RECONSTRUCTION: AN UNSUPERVISED LEARNING BASED APPROACH 26

LOW DOSE ABDOMINAL CT IMAGE RECONSTRUCTION: AN UNSUPERVISED ...

引用

26th IEEE International Conference on Image Processing (ICIP)

作者： Kuanar, Shiba Athitsos, Vassilis Mahapatra, Dwarikanath Rao, K. R. Akhtar, Zahid Dasgupta, Dipankar Univ Texas Arlington Dept Elect Arlington TX 76019 USA Univ Texas Arlington Dept Comp Sci Engn Arlington TX 76019 USA IBM Res Melbourne Vic Australia Univ Memphis Dept Comp Sci Memphis TN 38152 USA

ISBN: (纸本)9781538662496

In medical practice, the X-ray Computed tomography-based scans expose a high radiation dose and lead to the risk of prostate or abdomen cancers. On the other hand, the low-dose CT scan can reduce radiation exposure to the patient. But the reduced radiation dose degrades image quality for human perception, and adversely affects the radiologist's diagnosis and prognosis. In this paper, we introduce a GAN based auto-encoder network to de-noise the CT images. Our network first maps CT images to low dimensional manifolds and then restore the images from its corresponding manifold representations. Our reconstruction algorithm separately calculates perceptual similarity, learns the latent feature maps, and achieves more accurate and visually pleasing reconstructions. We also showed the effectiveness of our model on a number of patient abdomen CT images, and compare our results with existing deep learning and iterative reconstruction methods. Experimental results demonstrate that our model outperforms other state-of-the-art methods in terms of PSNR, SSIM, and statistical properties of the image regions.

关键词： auto-encoder Low dose CT Image De-noise Manifold GAN

来源：评论

学校读者我要写书评

暂无评论

Collaborative Generative Adversarial Network for Recommendation Systems 35

Collaborative Generative Adversarial Network for Recommendat...

引用

IEEE 35th International Conference on Data Engineering (ICDE)

作者： Tong, Yuzhen Luo, Yadan Zhang, Zheng Sadiq, Shazia Cui, Peng Univ Queensland Brisbane Qld Australia Tsinghua Univ Beijing Peoples R China

ISBN: (纸本)9781728108902

Recommendation systems have been a core part of daily Internet life. Conventional recommendation models hardly defend adversaries due to the natural noise like misclicking. Recent researches on GAN-based recommendation systems can improve the robustness of the learning models, yielding the state-of-the-art performance. The basic idea is to adopt an interplay minimax game on two recommendation systems by picking negative samples as fake items and employ reinforcement learning policy. However, such strategy may lead to mode collapse and result in high vulnerability to adversarial perturbations on its model parameters. In this paper, we propose a new collaborative framework, namely Collaborative Generative Adversarial Network (CGAN), which adopts Variational auto-encoder (VAE) as the generator and performs adversarial training in the continuous embedding space. The formulation of CGAN has two advantages: 1) its auto-encoder takes the role of generator to mimic the true distribution of users preferences over items by capturing subtle latent factors underlying user-item interactions;2) the adversarial training in continuous space enhances models robustness and performance. Extensive experiments conducted on two real-world benchmark recommendation datasets demonstrate the superior performance of our CGAN in comparison with the state-of-theart GAN-based methods.

关键词： Collaborative Filter Adversarial Training auto-encoder Recommendation Systems

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：