Improved GAN: Using a transformer module generator approach for material decomposition

Wang, Guoshuai; Liu, Zhou; Huang, Zhengyong; Zhang, Na; Luo, Honghong; Liu, Lijian; Shen, Hao; Che, Canwen; Niu, Tianye; Liang, Dong; Luo, Di; Hu, Zhanli

doi:10.1016/j.compbiomed.2022.105952

Cited by 10 publications

(12 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Likely, Y. Jiang, Chang, and Wang (2021), G. Wang et al (2022), B. Zhang, Gu, et al (2022 focused on the combination of GAN and transformers to stabilize GAN training and enhance the quality of the generated samples. Also, Zadorozhnyy et al (2021) proposed loss functions to modify the training gradient and support high discriminator scores for the real data.…”

Section: Training Instabilitymentioning

confidence: 99%

“…The authors' idea is based on taking advantage of the auto‐encoder to learn the discriminator with a reduced representation of the given data and getting the benefits of the auto‐encoder representative features and the discriminator's discriminative features to regularize the discriminator and then stabilize GAN training. Likely, Y. Jiang, Chang, and Wang (2021), G. Wang et al (2022), B. Zhang, Gu, et al (2022) focused on the combination of GAN and transformers to stabilize GAN training and enhance the quality of the generated samples. Also, Zadorozhnyy et al (2021) proposed loss functions to modify the training gradient and support high discriminator scores for the real data.…”

Section: Gan Challengesmentioning

confidence: 99%

See 1 more Smart Citation

A comprehensive review of generative adversarial networks: Fundamentals, applications, and challenges

Mohammed

2023

WIREs Computational Stats

View full text Add to dashboard Cite

In machine learning, a generative model is responsible for generating new samples of data in terms of a probabilistic model. Generative adversarial network (GAN) has been widely used to generate realistic samples in different domains and outperforms its peers in the generative models family. However, producing a robust GAN model is not a trivial task because many challenges face the GAN during the training process and impact its performance, affecting the quality and diversity of the generated samples. In this article, we conduct a comprehensive review of GANs to present the fundamentals of GAN, including its components, types, and objective functions. Also, we present an overview of the evaluation matrices used to evaluate GAN models. Moreover, we list the applications of GANs and research work in various domains. Finally, we present the challenges that face GANs and highlight two significant issues, representing mode collapse and training instability, in addition to those research efforts that tackle these challenges.This article is categorized under: Statistical Learning and Exploratory Methods of the Data Sciences > Deep Learning Statistical Learning and Exploratory Methods of the Data Sciences > Neural Networks

show abstract

Section: Training Instabilitymentioning

confidence: 99%

Section: Gan Challengesmentioning

confidence: 99%

A comprehensive review of generative adversarial networks: Fundamentals, applications, and challenges

Mohammed

2023

WIREs Computational Stats

View full text Add to dashboard Cite

show abstract

“…e optimal discriminator D * (x) is obtained by using the loss function in (12), fixing the generator G, and deriving it for the discriminator D(x), as shown in…”

Section: Data Discriminator Modulementioning

confidence: 99%

“…Compare the anomaly detection performance of DUAL-ADGAN with the other nine unsupervised anomaly detection baseline models on three datasets, RealAdExchange-CPC, RealTraffic-SPEED, and Real-Traffic-TravelTime. (3) For epochs do (4) Feed the noise vector z into the generator G W to generate the data G W (5) Feed the generated data G W (z) and the real data x into the discriminator D W (6) Training G W and D W with WGAN-GP loss function (7) Return G W (8) If model is Fence-GAN: (9) For epochs do (10) Feed the noise vector z into the generator G F to generate the data G F (z) (11) Feed the generated data G F (z) and the real data x into the discriminator D F (12) Training G F and D F with Fence-GAN loss function (13) Step 2.…”

Section: Experimental Protocolmentioning

confidence: 99%

“…With the development of generative adversarial networks in recent years, this kind of generative deep learning model by adversarial training method has received more and more attention from researchers. GAN [ 11 ] models are used to learn high-dimensional complex data distributions in the real world, and they have made great progress in image processing tasks such as medical image synthesis [ 12 ] and superresolution reconstruction. Meanwhile, the researchers found that mapping the data inverse to the random potential space and reconstructing the error according to the generator can be effective for anomaly detection tasks.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Research on DUAL-ADGAN Model for Anomaly Detection Method in Time-Series Data

Gong

Wang

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

In recent years, anomaly detection techniques in time-series data have been widely used in manufacturing, cybersecurity, and other fields. Meanwhile, various anomaly detection models based on generative adversarial networks (GAN) are gradually used in time-series anomaly detection tasks. However, there are problems of unstable generator training, missed detection of anomalous data, and inconsistency between the discriminator’s discriminant and the anomaly detection target in GAN networks. Aiming at the above problems, the paper proposes a DUAL-ADGAN (Dual Anomaly Detection Generative Adversarial Networks) model for the detection of anomalous data in time series. First, the Wasserstein distance satisfying the Lipschitz constraint is used as the loss function of the data reconstruction module, which improves the stability of the traditional GAN network training. Second, by adding a data prediction module to the DUAL-ADGAN model, the distinction between abnormal and normal samples is increased, and the rate of missing abnormal data in the model is reduced. Third, by introducing the Fence-GAN loss function, the discriminator is aligned with the anomaly detection target, which effectively reduces the anomaly data false detection rate of the DUAL-ADGAN model. Finally, anomaly scores derived from the DUAL-ADGAN model are compared with dynamic thresholds to detect anomalies. The experimental results show that the average F1 of the DUAL-ADGAN model is 0.881, which is better than the other nine baseline models. The conclusions demonstrate that the DUAL-ADGAN model proposed in the paper is more stable in training while effectively solving the problems of anomaly miss detection and discriminator inconsistency with the anomaly detection target in the anomaly detection task.

show abstract