denoising autoencoder (DAE) is effective in restoring clean speech from noisy observations. In addition, it is easy to be stacked to a deep denoising autoencoder (DDAE) architecture to further improve the performance....
详细信息
ISBN:
(纸本)9781634394352
denoising autoencoder (DAE) is effective in restoring clean speech from noisy observations. In addition, it is easy to be stacked to a deep denoising autoencoder (DDAE) architecture to further improve the performance. In most studies, it is supposed that the DAE or DDAE can learn any complex transform functions to approximate the transform relation between noisy and clean speech. However, for large variations of speech patterns and noisy environments, the learned model is lack of focus on local transformations. In this study, we propose an ensemble modeling of DAE to learn both the global and local transform functions. In the ensemble modeling, local transform functions are learned by several DAEs using data sets obtained from unsupervised data clustering and partition. The final transform function used for speech restoration is a combination of all the learned local transform functions. Speech denoising experiments were carried out to examine the performance of the proposed method. Experimental results showed that the proposed ensemble DAE model provided superior restoration accuracy than traditional DAE models.
In this paper, we propose a robust distant-talking speech recognition by combining cepstral domain denoising autoencoder (DAE) and temporal structure normalization (TSN) filter. For the proposed method, after applying...
详细信息
ISBN:
(纸本)9781479942190
In this paper, we propose a robust distant-talking speech recognition by combining cepstral domain denoising autoencoder (DAE) and temporal structure normalization (TSN) filter. For the proposed method, after applying a DAE in the cepstral domain of speech to suppress reverberation, we apply a postprocessing technology based on temporal structure normalization (TSN) filter to reduce the noise and reverberation effects by normalizing the modulation spectra to reference spectra of clean speech. The proposed method was evaluated using speech in simulated and real reverberant environments. By combining a cepstral-domain DAE and TSN, the average Word Error Rate (WER) was reduced from 25.2% of the baseline system to 21.2% in simulated environments and from 47.5% to 41.3% in real environments, respectively.
Microarray data analysis has emerged as a strong tool for cancer diagnosis. Nevertheless, researches on it are significantly challenging as the microarray datasets are imbalanced and high-dimensional with relatively s...
详细信息
ISBN:
(纸本)9781728114194
Microarray data analysis has emerged as a strong tool for cancer diagnosis. Nevertheless, researches on it are significantly challenging as the microarray datasets are imbalanced and high-dimensional with relatively small sample size. In this paper, we utilized Dual denoising autoencoder Features (DDAF), which integrates two denoising Auto-Encoders (DAE) with different activation function to map the features for both minority and majority classes into a better classification representation. The experimental results on four typical microarray datasets show that the DDAF outperforms the Dual autoencoder Features (DAF) and the Cost-sensitive Oversampling Stacked denoising Auto-Encoder (CO-SDAE), rendering the robust ability for dimensionality reduction and imbalanced classification.
Improving the performance of deep learning and making it more in line with real life have always been the research direction of artificial intelligence. In this paper, a denoising autoencoder genetic algorithm convolu...
详细信息
ISBN:
(纸本)9789881563958
Improving the performance of deep learning and making it more in line with real life have always been the research direction of artificial intelligence. In this paper, a denoising autoencoder genetic algorithm convolution neural network (DGCNN) model based on deep learning is proposed. Two parts of the research work are combined to improve its performance. Firstly, the traditional autoencoder is replaced by the denoising autoencoder and improved the way of adding noise. Secondly, genetic algorithm is utilized to combine CNN at the stage of image classification. This allows DGCNN to cope with complex and volatile situations while enhancing image processing capabilities. Simulation results show that the method can enhance the ability of image processing than traditional methods. The performance of the proposed model is better than traditional method when the images of different loss levels are processed by this method. The results are verifying the feasibility and effectiveness of the model and algorithm. DGCNN shows better capacity in improving the performance of image processing and dealing with complex situations effectively.
Phishing is referred as an attempt to obtain sensitive information, such as usernames, passwords, and credit card details (and, indirectly, money), for malicious reasons, by disguising as a trustworthy entity in an el...
详细信息
ISBN:
(纸本)9781450354301
Phishing is referred as an attempt to obtain sensitive information, such as usernames, passwords, and credit card details (and, indirectly, money), for malicious reasons, by disguising as a trustworthy entity in an electronic communication [1]. Hackers and malicious users, often use Emails as phishing tools to obtain the personal data of legitimate users, by sending Emails with authentic identities, legitimate content, but also with malicious URL, which help them to steal consumer's data. The high dimensional data in phishing context contains large number of redundant features that significantly elevate the classification error. Additionally, the time required to perform classification increases with the number of features. So extracting complex Features from phishing Emails requires us to determine which Features are relevant and fundamental in phishing detection. The dominant approaches in phishing are based on machine learning techniques;these rely on manual feature engineering, which is time consuming. On the other hand, deep learning is a promising alternative to traditional methods. The main idea of deep learning techniques is to learn complex features extracted from data with minimum external contribution [2]. In this paper, we propose new phishing detection and prevention approach, based first on our previous spam filter [3] to classify textual content of Email. Secondly it's based on autoencoder and on denoising autoencoder (DAE), to extract relevant and robust features set of URL (to which the website is actually directed), therefore the features space could be reduced considerably, and thus decreasing the phishing detection time.
Music embedding often causes significant performance degradation in automatic speech recognition (ASR). This paper proposes a music-removal method based on denoising autoencoder (DAE) that learns and removes music fro...
详细信息
ISBN:
(纸本)9789881476807
Music embedding often causes significant performance degradation in automatic speech recognition (ASR). This paper proposes a music-removal method based on denoising autoencoder (DAE) that learns and removes music from music-embedded speech signals. Particularly, we focus on convolutional denoising autoencoder (CDAE) that can learn local musical patterns by convolutional feature extraction. Our study shows that the CDAE model can learn patterns of music in different genres and the CDAE-based music removal offers significant performance improvement for ASR. Additionally, we demonstrate that this music-removal approach is largely language independent, which means that a model trained with data in one language can be applied to remove music from speech in another language, and models trained with multilingual data may lead to better performance.
Rogue emitter detection (RED) is a crucial technique to maintain secure internet of things applications. Existing deep learning-based RED methods have been proposed under friendly environments. However, these methods ...
详细信息
ISBN:
(纸本)9781538674628
Rogue emitter detection (RED) is a crucial technique to maintain secure internet of things applications. Existing deep learning-based RED methods have been proposed under friendly environments. However, these methods perform unstably under low signal-to-noise ratio (SNR) scenarios. To address this problem, we propose a robust RED method, which is a hybrid network of denoising autoencoder and deep metric learning (DML). Specifically, denoising autoencoder is adopted to mitigate noise interference and then improve its robustness under low SNR while DML plays an important role to improve the feature discrimination. Several typical experiments are conducted to evaluate the proposed RED method on an automatic dependent surveillance-Broadcast dataset and an IEEE 802.11 dataset and also to compare it with existing RED methods. Simulation results show that the proposed method achieves better RED performance and higher noise robustness with more discriminative semantic vectors than existing methods.
Credit risk evaluation is a key consideration in financial activities. Financial institutions such as banks rely on credit risk analysis for determining the potential risk involved in financial activities and then dec...
详细信息
ISBN:
(纸本)9781450364195
Credit risk evaluation is a key consideration in financial activities. Financial institutions such as banks rely on credit risk analysis for determining the potential risk involved in financial activities and then decide the degree of involvement in such activities as well as the appropriate interest rate and the amount of capital that should be reserved. The recent development of machine learning has provided powerful tools for computer-aided credit risk analysis, and neural networks are one of the most promising approaches. However, conventional artificial neural networks involve multiple layers of neurons which then become a universal function that can approximate any function. Therefore, it will learn from not only the information in the training data set but also from the noise in it. It is critical to remove the noise in order to improve the accuracy and efficiency of such algorithms. In this paper, a denoising autoencoder approach is proposed for the training process for neural networks. The denoising-autoencoder-based neural network model is then applied to credit risk analysis, and the performance is evaluated.
The network connection of the power cyber-physical system (PCPS) makes it easy to become a potential attack target, leading to serious consequences such as paralysis of the power system. The existence of class imbalan...
详细信息
ISBN:
(纸本)9789819756056;9789819756063
The network connection of the power cyber-physical system (PCPS) makes it easy to become a potential attack target, leading to serious consequences such as paralysis of the power system. The existence of class imbalance and noise in the PCPS network traffic dataset limits the intrusion detection accuracy. In response to the above difficulties, this article combines the denoising autoencoder and the CNN-Attention network to define the AE-CNA NIDS architecture. The BatchSwapNoise method DAE is used to denoise, compress and reconstruct features of the data, and extract local spatiotemporal features in theCNAblock formed by CNN and Attention. Multiple CNA blocks are stacked together to comprehensively learn the multi-layer spatiotemporal characteristics of network attack data. In addition, for the network intrusion dataset imbalance problem, equalization loss v2 (EQL v2) is used to balance the weight attention of the minority class. Experimental results show that AE-CNA performs well in terms of accuracy, precision, and recall value, with an accuracy rate of 98.3%, effectively improving intrusion detection performance and minority class detection rate.
作者：
Zhang, JianyeYin, PengChinese Acad Sci
Joint Engn Res Ctr Hlth Big Data Intelligent Anal Shenzhen Inst Adv Technol Shenzhen Peoples R China Tsinghua Univ
Shenzhen Int Grad Sch Dept Comp Sci & Technol Shenzhen Peoples R China
This paper presents a novel method for imputing missing data of multivariate time series by adapting the Long Short Term-Memory(LSTM) and denoising autoencoder(DAE). Missing data are ubiquitous in many domains;proper ...
详细信息
ISBN:
(纸本)9781728118673
This paper presents a novel method for imputing missing data of multivariate time series by adapting the Long Short Term-Memory(LSTM) and denoising autoencoder(DAE). Missing data are ubiquitous in many domains;proper imputation methods can improve performance on many tasks. Our method focus on multivariate time series, applying bidirectional LSTM to learn temporal information and DAE to learn correlation between variables, and we combine these two models by using LSTM as the encoder component of DAE. Several real-world datasets, including electroencephalogram(EEG), electromyogram(EMG) and electronic health records(EHRs), are extracted to test the performance of our method. Through simulation studies, we compare the proposed recurrent denoising autoencoder with several baseline imputation methods and demonstrate its effectiveness in both missing data estimation and label prediction after imputation.
暂无评论