The excellent performance of representation learning of autoencoders have attracted considerable interest in various applications. However, the structure and multi-local collaborative relationships of unlabeled data a...
详细信息
The excellent performance of representation learning of autoencoders have attracted considerable interest in various applications. However, the structure and multi-local collaborative relationships of unlabeled data are ignored in their encoding procedure that limits the capability of feature extraction. This paper presents a Multi-local Collaborative autoencoder (MC-AE), which consists of novel multi local collaborative representation RBM (mcrRBM) and multi-local collaborative representation GRBM (mcrGRBM) models. Here, the Locality Sensitive Hashing (LSH) method is used to divide the input data into multi-local cross blocks which contains multi-local collaborative relationships of the unlabeled data and features since the similar multi-local instances and features of the input data are divided into the same block. In mcrRBM and mcrGRBM models, the structure and multi-local collaborative relationships of unlabeled data are integrated into their encoding procedure. Then, the local hidden features converges on the center of each local collaborative block. Under the collaborative joint influence of each local block, the proposed MC-AE has powerful capability of representation learning for unsupervised clustering. However, our MC-AE model perhaps perform training process for a long time on the large-scale and high-dimensional datasets because more local collaborative blocks are integrate into it. Five most related deep models are compared with our MC-AE. The experimental results show that the proposed MC-AE has more excellent capabilities of collaborative representation and generalization than the contrastive deep models. (c) 2021 Elsevier B.V. All rights reserved.
Deep convolutional autoencoder (DCAE) is usually optimized to minimize the difference between the input and the reconstruction, and the reconstruction error has been widely used as an indicator for visual anomaly dete...
详细信息
ISBN:
(纸本)9781665441155
Deep convolutional autoencoder (DCAE) is usually optimized to minimize the difference between the input and the reconstruction, and the reconstruction error has been widely used as an indicator for visual anomaly detection. However, DCAE sometimes can reconstruct anomalies very well and thus may yield misdetections. To tackle this issue, we propose a novel non-symmetrical DCAE, which is trained in a two-stage manner. Specifically, a single RotNet is first trained to serve as encoder. Then, discriminative representations generated by the frozen encoder are used to train two parallel decoders for image reconstruction. Finally, the reconstruction errors obtained by the two decoders are combined as the anomaly score. Massive experiments on three public datasets and one practical industrial dataset demonstrate the superiority of the proposed method among existing reconstruction based methods.
In this paper, a deep learning approach is introduced to detect pathological voice disorders from continuous speech. Speech as bio-signal is getting more and more attention as a discriminant for different diseases. To...
详细信息
ISBN:
(纸本)9789897584909
In this paper, a deep learning approach is introduced to detect pathological voice disorders from continuous speech. Speech as bio-signal is getting more and more attention as a discriminant for different diseases. To exploit information in speech, a long-short term memory (LSTM) autoencoder hybrid with multi-task learning solution is proposed with spectrogram as input feature. Different speech databases (voice disorders, depression, Parkinson's disease) are applied as evaluation datasets. Applicability of the method is demonstrated by obtaining accuracies 85% for Parkinson's disease, 86% for dysphonia, and 90% for depression on test datasets. The advantage of this method is that it is fully data-driven, in the sense that it does not require special acoustic-phonetic preprocessing separately for the types of disease to be recognized. We believe that the applied method in this article can be used to other diseases as well and can be used for other languages also.
Visual crypto-system is a class of cryptography intended to secure images. Random-grid crypto-system is a type of visual cryptosystem that generates an encrypted grid of the secret image utilizing a pre-encoded grid a...
详细信息
Visual crypto-system is a class of cryptography intended to secure images. Random-grid crypto-system is a type of visual cryptosystem that generates an encrypted grid of the secret image utilizing a pre-encoded grid and the secret image. The random grid research is still engaging in three dimensions: security, quality, and efficiency. Though there are many works in improving security, there is scope for investigation in the other two directions from the perspective of technological advancements. There has been a significant increase in the number of Graphical Processing Unit (GPU) cores for which the random grid models are intuitively amenable. The random grid secret sharing models demand more improvement in the quality of the reconstructed image as they achieved only 50% contrast. In this paper, we proposed a GPU based random-grid model to improve its efficiency by exploiting the data-parallelism inherent in the model. In addition to this speedup of 3151x, we restored the secret image with a quality almost equal to the original secret image using autoencoder super-resolution. Objective quality measures such as MSE, NCC, NAE and SSIM for the proposed model empirically confirm the improvement in image quality compared to other state-of-the-art models.
In the past decade,recommender systems have been widely used to provide users with personalized products and ***,most traditional recommender systems are still facing a challenge in dealing with the huge volume,comple...
详细信息
In the past decade,recommender systems have been widely used to provide users with personalized products and ***,most traditional recommender systems are still facing a challenge in dealing with the huge volume,complexity,and dynamics of *** tackle this challenge,many studies have been conducted to improve recommender system by integrating deep learning *** an unsupervised deep learning method,autoencoder has been widely used for its excellent performance in data dimensionality reduction,feature extraction,and data ***,recent researches have shown the high efficiency of autoencoder in information retrieval and recommendation *** autoencoder on recommender systems would improve the quality of recommendations due to its better understanding of users,demands and characteristics of *** paper reviews the recent researches on autoencoder-based recommender *** differences between autoencoder-based recommender systems and traditional recommender systems are presented in this *** last,some potential research directions of autoencoder-based recommender systems are discussed.
Dynamic community detection is significant for controlling and capturing the temporal features of networks. The evolutionary clustering framework provides a temporal smoothness constraint for simultaneously maximizing...
详细信息
Dynamic community detection is significant for controlling and capturing the temporal features of networks. The evolutionary clustering framework provides a temporal smoothness constraint for simultaneously maximizing the clustering quality at the current time step and minimizing the clustering deviation between two successive time steps. Based on this framework, some existing methods, such as the evolutionary spectral clustering and evolutionary nonnegative matrix factorization, aim to look for the low-dimensional representation by mapping reconstruction. However, such reconstruction does not address the nonlinear characteristics of networks. In this paper, we propose a semi-supervised algorithm(sE-autoencoder) to overcome the effects of nonlinear property on the low-dimensional representation. Our proposed method extends the typical nonlinear reconstruction model to the dynamic network by constructing a temporal matrix. More specifically, the potential community characteristics and the previous clustering, as the prior information,are incorporated into the loss function as a regularization term. Experimental results on synthetic and realworld datasets demonstrate that the proposed method is effective and superior to other methods for dynamic community detection.
Health indicator (HI) construction is the most significant task of degradation assessment (DA) that facilitates prognostic and health management of rotating machinery. Many stacked autoencoder (SAE) models represented...
详细信息
Health indicator (HI) construction is the most significant task of degradation assessment (DA) that facilitates prognostic and health management of rotating machinery. Many stacked autoencoder (SAE) models represented by CNN-based and RNN-based SAE have been applied to the field of DA. However, the former has a small receptive vision which makes it weak in encoding time-series information, while the latter can easily encounter the problem of overfitting or parameter expansion. To solve these problems, this paper proposes an embedded LSTM-CNN autoencoder to extract trend features that contain both local characteristics and degradation trend information from vibration data. And, a transfer learning-based two-phase network training algorithm is designed to enhance the ability of noise filtering of the model. Then, HI is obtained by fusing the extracted trend features with a growing self-organized map. Finally, two case studies are implemented by using bearing datasets to verify the proposed method. The results show that HI gained by the proposed method is more effective than that by other existing methods. Moreover, the goodness-of-fit of polynomial degradation models with the HI is analyzed. (c) 2022 Elsevier B.V. All rights reserved.
Unmanned surface vehicle(USV)is currently a hot research topic in maritime communication network(MCN),where denoising and semantic segmentation of maritime images taken by USV have been rarely *** former has recently ...
详细信息
Unmanned surface vehicle(USV)is currently a hot research topic in maritime communication network(MCN),where denoising and semantic segmentation of maritime images taken by USV have been rarely *** former has recently researched on autoencoder model used for image denoising,but the existed models are too complicated to be suitable for real-time detection of *** this paper,we proposed a lightweight autoencoder combined with inception module for maritime image denoising in different noisy environments and explore the effect of different inception modules on the denoising ***,we completed the semantic segmentation task for maritime images taken by USV utilizing the pretrained U-Net model with tuning,and compared them with original U-Net model based on different ***,we compared the semantic segmentation of noised and denoised maritime images respectively to explore the effect of image noise on semantic segmentation *** studies are provided to prove the feasibility of our proposed denoising and segmentation ***,a simple integrated communication system combining image denoising and segmentation for USV is shown.
According to the smart manufacturing paradigm, the analysis of assets' time series with a machine learning approach can effectively prevent unplanned production downtimes by detecting assets' anomalous operati...
详细信息
According to the smart manufacturing paradigm, the analysis of assets' time series with a machine learning approach can effectively prevent unplanned production downtimes by detecting assets' anomalous operational conditions. To support smart manufacturing operators with no data science background, we propose an anomaly detection approach based on deep learning and aimed at providing a manageable machine learning pipeline and easy to interpret outcome. To do so we combine (i) an autoencoder, a deep neural network able to produce an anomaly score for each provided time series, and (ii) a discriminator based on a general heuristics, to automatically discern anomalies from regular instances. We prove the convenience of the proposed approach by comparing its performances against isolation forest with different case studies addressing industrial laundry assets' power consumption and bearing vibrations. (C) 2020 Elsevier B.V. All rights reserved.
Given the uniqueness of synthetic aperture radar (SAR) images, traditional optical image compression algorithms cannot fully exploit their redundant information. To improve SAR image compression in terms of rate-disto...
详细信息
Given the uniqueness of synthetic aperture radar (SAR) images, traditional optical image compression algorithms cannot fully exploit their redundant information. To improve SAR image compression in terms of rate-distortion performance and visual perception, an end-to-end SAR image compression convolutional neural network (CNN) model based on a variational autoencoder is proposed. The proposed CNN model consists of a main autoencoder and a hyper autoencoder. To reduce dependencies in latent space, a joint transform of linear CNN and nonlinear generalized divisive normalization (GDN) activation is applied in the main autoencoder. Moreover, residual blocks are combined with the transforms to boost the efficiency of feature learning and make use of subpixels to improve the quality of reconstructed images. Instead of a fixed entropy model, a conditioned entropy model that works with a hyperprior network is used to learn the distribution of latents, which helps to further improve the compression quality. During training, the model is optimized by evaluating the rate-distortion performance. The experimental results show that the proposed method can achieve better distortion performance than JPEG, JPEG2000, and the available CNN-based method in terms of objective evaluation criteria and human vision perception quality.
暂无评论