Zero-shot learning recognizes the unseen samples via the model learned from the seen class samples and semantic features. Due to the lack of information of unseen class samples in the training set, some researchers ha...
详细信息
Zero-shot learning recognizes the unseen samples via the model learned from the seen class samples and semantic features. Due to the lack of information of unseen class samples in the training set, some researchers have proposed the method of generating unseen class samples by using generative models. However, the generated model is trained with the training set samples first, and then the unseen class samples are generated, which results in the features of the unseen class samples tending to be biased toward the seen class and may produce large deviations from the real unseen class samples. To tackle this problem, we use the autoencoder method to generate the unseen class samples and combine the semantic features of the unseen classes with the proposed new sample features to construct the loss function. The proposed method is validated on three datasets and showed good results.
Deep learning has been developed to generate promising super resolution hyperspectral imagery by fusing hyperspectral imagery with the panchromatic ***,it is still challenging to maintain edge spectral information in ...
详细信息
Deep learning has been developed to generate promising super resolution hyperspectral imagery by fusing hyperspectral imagery with the panchromatic ***,it is still challenging to maintain edge spectral information in the necessary upsampling processes of these approaches,and diffcult to guarantee effective feature *** study proposes a pansharpening network denoted as HyperRefiner that consists of,(1)a well performing upsampling network SRNet,in which the dual attention block and refined attention block are cascaded to accomplish the extraction and fusion of features;(2)a spectral autoencoder that is embedded to perform dimensionality reduction under constrained feature extraction;and(3)the optimization module which performs self-attention at the pixel and feature levels.A comparisonwithseveral state-of-the-art models reveals that HyperRefiner can improve the quality of the fused ***,compared to the single-head HyperTransformer and with the Chikusei dataset,our network improved the Peak Signal-to-Noise Ratio,Erreur Relative Globale Adimensionnelle de Synthese and Spectral Angle Mapper by 0.86%,3.62%,and 2.09%,and reduce the total memory,floating point operations,model parameters and computation time by 41%,75%,86%and 46%,*** experimental results show that HyperRefiner outperforms several networks and demonstrates its usefulness in hyperspectral image *** code is publicly available athttps://***/zsspo/Fusion_HyperRefiner.
Images occupy a prominent place in data because they are more visually appealing than sounds or texts. Acoustic waves are the only feasible alternative for long-distance underwater transmission since seawater has a hi...
详细信息
Images occupy a prominent place in data because they are more visually appealing than sounds or texts. Acoustic waves are the only feasible alternative for long-distance underwater transmission since seawater has a high absorption impact on lighting and electromagnetic signals. However, underwater acoustic (UWA) communication technology can only provide relatively limited bandwidth (low effectiveness) and insufficiently stable links (low reliability). As a result, it is challenging to send high-resolution underwater images via the UWA channel. This article proposes an effective and robust underwater image compression scheme. First, an autoencoder is used for underwater image extreme bit rate compression. Then, a multistep training strategy is proposed to improve the robustness of the decoder by gradually learning channel degradation features. Finally, the autoencoder encodes images in two paths to achieve efficient compression and higher image quality when reconstructing images. The main path completes the compression task with a low bit rate and high robustness, while the branching path implements the image block retransmission compensation through the feedback signal. The experimental results demonstrate that the content of the reconstructed image can still be recognized under the conditions of a compression ratio of up to 1/768 and an average bit error rate of up to 10(-1). The joint multistep training strategy and multidescription coding achieve a low bit rate and high robustness for underwater image communication, which has good application prospects.
Detecting anomalies such as breakage and excessive wear of cutting tools in the machining process is crucial to prevent damage and improve productivity. Data-driven anomaly detection (AD) methods suffer from limited a...
详细信息
Detecting anomalies such as breakage and excessive wear of cutting tools in the machining process is crucial to prevent damage and improve productivity. Data-driven anomaly detection (AD) methods suffer from limited availability of anomaly samples, which is ineluctable in practice owing to strict reliability restrictions. Therefore, we propose a semisupervised AD approach in which only failure-free samples are required to establish an AD model. The key strategy is to learn the characteristics of failure-free samples using an improved autoencoder (AE) and discern observations by deviations from the characteristics. We rebuild the loss function of AE to impel the model to learn the common characteristics in latent space. We propose a factor that reflects the anomaly degree as the decision-making function to implement AD. The proposed approach is verified on an experimental cutting tool breakage dataset and a public cutting tool wear dataset. The experimental results demonstrate the validity of the proposed approach. The comparisons with conventional methods substantiate that the proposed approach outperforms existing AD methods.
Nowadays, Deep learning (DL) techniques have been proven successful as learning techniques in various research fields ranging from computer vision to social networks. The approach of DL is flourishing in the field of ...
详细信息
Nowadays, Deep learning (DL) techniques have been proven successful as learning techniques in various research fields ranging from computer vision to social networks. The approach of DL is flourishing in the field of recommender systems (RS). Researchers have deployed metadata or auxiliary information using DL approaches in diverse applications in the last decade to achieve better recommendation accuracy. Thus, the metadata plays a vital role in obtaining a better user-item interaction. At the same time, existing techniques are based on fixed user and item factors. Therefore, the model does not correctly identify actual latent factors representation, resulting in a high prediction error. To handle this problem, a user metadata embedding using a deep autoencoder RS model called "Metadata Embedding Deep autoencoder (MEDAE)" based collaborative filtering is proposed. MEDAE model takes embeds user metadata such as demographics along with the rating data. The MEDAE model consists of an embedding layer, Encoder, and Decoder. The embedding layer generates embedding or latent features of the users, items, and metadata;Encoder receives concatenated features of the user, item, and metadata, then encodes the inputs and passes them to the decoder;and the decoder reconstructs the output. To test the effectiveness of proposed model Root Mean Squared Error and Mean Absolute Error measures are used. Different architectures (like Big-Small-Big (BSB) (5), BSB (3), Small-Big-Small (3), and SBS (5)) of the MEDAE model are evaluated on MovieLens datasets along with different parameters such as activation functions (ELU and SELU) and regularization and results concluded that the MEDAE with SBS (3) and ELU + SELU component improves 4% of RMSE and 2% MAE over the baseline methods.
The use of flying Unmanned Aerial Vehicles (UAVs) for communications is becoming more and more widespread, especially in 5G and beyond networks. In such a context, detection and authentication of UAVs is assuming an i...
详细信息
ISBN:
(纸本)9798350311143
The use of flying Unmanned Aerial Vehicles (UAVs) for communications is becoming more and more widespread, especially in 5G and beyond networks. In such a context, detection and authentication of UAVs is assuming an increasingly important role. In this paper we show that it is possible to distinguish different drones which communicate with a fixed ground base station (BS) on the basis of their channel characteristics and of the micro-Doppler signature associated to the specific features of each UAV. An urban scenario is simulated where UAVs fly at a constant height and channels are affected by Additive White Gaussian Noise (AWGN) and fading. With the aim of helping the BS in its authentication task, we take advantage of a sparse autoencoder trained on the channel of the legitimate transmitter, while data coming from possible attackers are classified as anomalies. We prove that, with proper network training, low levels of false alarm and missed detection can be achieved, especially if the attacker has no line-of-sight link, and that the presence of micro-Doppler actually contribute to enhance the authentication performance.
Passive space-borne radiometers operating in the 1400-1427 MHz protected frequency band face radio frequency interference (RFI) from terrestrial sources. With the growth of wireless devices and the appearance of new t...
详细信息
ISBN:
(纸本)9798350320107
Passive space-borne radiometers operating in the 1400-1427 MHz protected frequency band face radio frequency interference (RFI) from terrestrial sources. With the growth of wireless devices and the appearance of new technologies, the possibility of sharing this spectrum with other technologies would introduce more RFI to these radiometers. This band could be an ideal mid-band frequency for 5G and Beyond, as it offers high capacity and good coverage. Current RFI detection and mitigation techniques at SMAP (Soil Moisture Active Passive) depend on correctly detecting and discarding or filtering the contaminated data leading to the loss of valuable information, especially in severe RFI cases. In this paper, we propose an autoencoder-based RFI mitigation method to remove the dominant RFI caused by potential coexistent terrestrial users (i.e., 5G base station) from the received contaminated signal at the passive receiver side, potentially preserving valuable information and preventing the contaminated data from being discarded (1).
A hard landing is a typical accident that occurs during the landing of an aircraft. Because hard landings can cause stress buildup in the aircraft structure that can lead to fatal accidents if not properly identified,...
详细信息
A hard landing is a typical accident that occurs during the landing of an aircraft. Because hard landings can cause stress buildup in the aircraft structure that can lead to fatal accidents if not properly identified, it is necessary to clearly determine whether a hard landing has occurred. However, erroneous judgments may occur because of the limitations of the existing methods of identifying hard landings. Although many studies have been conducted to reduce misjudgments, most existing approaches require the selection of proper key (flight) parameters or predefined thresholds, which require a great deal of experience and high-level professional knowledge. Therefore, in this study, a new model for identifying hard landings without explicit selection of key parameters or manual determination of thresholds is proposed by introducing an outlier detection technique. An autoencoder, an artificial neural network model, is applied to the detection of outliers from landing data obtained through high-fidelity landing simulation. The training of the autoencoder and performance analysis is conducted to demonstrate the validity of the proposed method. Normal landing data are used as the training dataset for the autoencoder, and the percentage of abnormal landing data are gradually increased to the training dataset to check the robustness of the proposed method. The performance analysis results showed that the proposed method applying the autoencoder can be successfully used to identify hard landing situations such as late flares, even if a small amount of abnormal data are included in the training dataset, as is the case for actual landing data.
As the maritime industry continues to thrive and maritime services diversify, the demand for highly reliable maritime communication systems has become increasingly prominent. However, harsh marine conditions pose sign...
详细信息
As the maritime industry continues to thrive and maritime services diversify, the demand for highly reliable maritime communication systems has become increasingly prominent. However, harsh marine conditions pose significant challenges to communication systems. In this work, we propose a Maritime autoencoder (MAE) communication system based on Attention Mechanisms (AMs) and DenseBlock (namely AM-Dense-MAE). AM-Dense-MAE utilizes DenseBlock and long short-term memory to extract deep features and capture spatio-temporal relationships, addressing the issue of "long-term dependency". Furthermore, the decoder incorporates spatial attention modules and convolutional block attention module to enhance the preservation of crucial information and suppress irrelevant data. We employ the Rician fading channel model to simulate maritime communication channels. A substantial volume of data is utilized for model training and parameter optimization. Simulation results demonstrate that, in comparison to the benchmarks, the proposed AM-Dense-MAE exhibits better block error rate performance under various signal-to-noise ratio conditions and showcases generalization capabilities across diverse parameter settings.
Popularity bias is a massive challenge for autoencoder-based models, which decreases the level of personalization and hurts the fairness of recommendations. User reviews reflect their preferences and help mitigate bia...
详细信息
Popularity bias is a massive challenge for autoencoder-based models, which decreases the level of personalization and hurts the fairness of recommendations. User reviews reflect their preferences and help mitigate bias or unfairness in the recommendation. However, most existing works typically incorporate user (item) reviews into a long document and then use the same module to process the document in parallel. Actually, the set of user reviews is completely different from the set of item reviews. User reviews are heterogeneous in that they reflect a variety of items purchased by users, while item reviews are only related to the item itself and are thus typically homogeneous. In this article, a novel asymmetric attention network fused with autoencoders is proposed, which jointly learns representations from the user and item reviews and implicit feedback to perform recommendations. Specifically, we design an asymmetric attentive module to capture rich representations from user and item reviews, respectively, which solves data sparsity and explainable problems. Furthermore, to further address popularity bias, we apply a noise-contrastive estimation objective to learn high-quality "de-popularity" embedding via the decoder structure. A series of extensive experiments are conducted on four benchmark datasets to show that leveraging user review information can eliminate popularity bias and improve performance compared to various state-of-the-art recommendation techniques.
暂无评论