Auto-Encoders in Deep Learning—A Review with New Perspectives

Chen, Shuangshuang; Wei, Guo

doi:10.3390/math11081777

Cited by 91 publications

(19 citation statements)

References 172 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Feature extraction is performed by employing two distinct encoder models for both the source and target datasets. These encoder models transform the input feature data into lowerdimensional representations [22]. It takes an input data vector 𝑥 with 𝑛 features, represented as…”

Section: Feature Extractionmentioning

confidence: 99%

Heterogeneous Cross-Project Defect Prediction Using Encoder Networks and Transfer Learning

Haque,

Ali,

McClean

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Heterogeneous cross-project defect prediction (HCPDP) aims to predict defects in new software projects using defect data from previous software projects where the source and target projects have some different metrics. Most existing methods only find linear relationships in the software defect features and datasets. Additionally, these methods use multiple defect datasets from different projects as source datasets. In this paper, we propose a novel method called heterogeneous cross-project defect prediction using encoder and transfer learning (ETL). ETL uses encoders to extract the important features from source and target datasets. Also, to minimize negative transfer during transfer learning, we used an augmented dataset that contains pseudo-labels and the source dataset. Additionally, we have used very limited data to train the model. To evaluate the performance of the ETL approach, 16 datasets from four publicly available software defect projects were used. Furthermore, we compared the proposed method with four HCPDP methods namely EGW, HDP_KS, CTKCCA and EMKCA, and one WPDP method from existing literature. The proposed method on average outperforms the baseline methods in terms of PD, PF, F1-score, G-mean and AUC.

show abstract

Section: Feature Extractionmentioning

confidence: 99%

Heterogeneous Cross-Project Defect Prediction Using Encoder Networks and Transfer Learning

Haque,

Ali,

McClean

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The architecture of the deep autoencoder [21] comprises 10 encoding layers and 10 decoding layers. Each encoding layer takes as input the output from the previous layer, enabling the learning of increasingly intricate features as information progresses through the network.…”

Section: Features Extraction With Autoencodermentioning

confidence: 99%

Intelligent intrusion detection through deep autoencoder and stacked long short-term memory

Moukhafi,

Tantaoui,

Chana

et al. 2024

IJECE

View full text Add to dashboard Cite

In the realm of network intrusion detection, the escalating complexity and diversity of cyber threats necessitate innovative approaches to enhance detection accuracy. This study introduces an integrated solution leveraging deep learning techniques for improved intrusion detection. The proposed framework consists on a deep autoencoder for feature extraction, and a stacked long short-term memory (LSTM) network ensemble for classification. The deep autoencoder compresses raw network data, extracting salient features and mitigating noise. Subsequently, the stacked LSTM ensemble captures intricate temporal dependencies, correcting anomaly detection precision. Experiments conducted on the UNSW-NB15 dataset, and a benchmark in intrusion detection validate the effectiveness of the approach. The solution achieves an accuracy of 90.59%, with precision, recall, and F1-Score metrics reaching 90.65, 90.59, and 90.57, respectively. Notably, the framework outperforms standalone models and demonstrates the advantage of synergizing deep autoencoder-driven feature extraction with the stacked LSTM ensemble. Furthermore, a binary classification experiment attains an accuracy of about 90.59%, surpassing the multiclass classification and affirming the model's potential for binary threat identification. Comparative analyses highlight the pivotal role of feature extraction, while experimentation illustrates the enhancement achieved by incorporating the synergistic deep autoencoder-Stacked LSTM approach.

show abstract

“…The autoencoder, consisting of an encoder and a decoder, is an unsupervised deep neural network that learns how to efficiently compress input data into a meaningful representation and subsequently reconstruct the original data from this compressed form (Chen & Guo, 2023). By connecting the encoder and decoder, the autoencoder effectively captures important patterns and variations present in the data, enabling comprehensive analysis and interpretation (Chen & Guo, 2023). In this study, an autoencoder neural network (figure 3) was built to analyze the measured VWC time series at 20 cm depth for the 9 sites.…”

Section: Autoencoder Neural Networkmentioning

confidence: 99%

Prediction of Hysteretic Matric Potential Dynamics Using Artificial Intelligence: Application of Autoencoder Neural Networks

Aqel,

Reusser,

Margreth

et al. 2024

Preprint

View full text Add to dashboard Cite

Abstract. Information on soil water potential is essential to assess soil moisture state, to prevent soil compaction in weak soils, and to optimize crop management. In lack of direct measurements, the soil water potential values must be deduced from soil water content dynamics that can be monitored at plot scale or obtained at larger scale from remote sensing information. Because the relationship between water content and soil water potential in natural field soils is highly ambiguous, the prediction of soil water potential from water content data is a big challenge. The hysteretic relationship observed in nine soil profiles in the region of Solothurn (Switzerland) is not a simple function of texture or wetting and drainage cycles but depends on seasonal patterns that may be related to soil structural dynamics. Because the physical mechanisms governing seasonal hysteresis are unclear, we developed a deep neural network model that predicts water potential changes using rainfall, potential evapotranspiration, and water content time series as inputs. To adapt the model for multiple locations, we incorporated a Deep Autoencoder Neural Network as a classifier. The autoencoder compresses the water content time series into a site-specific feature that is highly representative of the underlying water content dynamics of each site and quantifies the similarity of dynamic patterns. By adding the Autoencoder's output as an additional input and training the neural network model with three stations located in three major classes founded by the autoencoder, we predict matric potential for other sites. This method has the potential to deduce the dynamics of matric potential from water content data (including satellite data) despite strong seasonal effects that cannot be captured by standard methods.

show abstract

Auto-Encoders in Deep Learning—A Review with New Perspectives

Cited by 91 publications

References 172 publications

Heterogeneous Cross-Project Defect Prediction Using Encoder Networks and Transfer Learning

Heterogeneous Cross-Project Defect Prediction Using Encoder Networks and Transfer Learning

Intelligent intrusion detection through deep autoencoder and stacked long short-term memory

Prediction of Hysteretic Matric Potential Dynamics Using Artificial Intelligence: Application of Autoencoder Neural Networks

Contact Info

Product

Resources

About