Article
Engineering, Multidisciplinary
Waad Almasri, Dimitri Bettebghor, Faouzi Adjed, Florence Danglade, Fakhreddine Ababsa
Summary: This study integrates layout and mechanical constraints in the mechanical design process using deep learning technology and proposes a DL-layout-driven solution trained via a generative adversarial network framework. The solution can quickly generate mechanically valid designs conforming with layout constraints and has the capability to generate multiple shapes based on different input constraints.
ENGINEERING OPTIMIZATION
(2022)
Article
Mechanics
Haoliang Jiang, Zhenguo Nie, Roselyn Yeo, Amir Barati Farimani, Levent Burak Kara
Summary: The use of deep learning for analyzing mechanical stress distributions has shown promising results, but most studies are limited in their applicability. We introduce a cGAN model called StressGAN that can predict stress distributions more accurately in various complex scenarios.
JOURNAL OF APPLIED MECHANICS-TRANSACTIONS OF THE ASME
(2021)
Article
Chemistry, Multidisciplinary
Hiskias Dingeto, Juntae Kim
Summary: While Machine Learning has security flaws, this paper proposes a Universal Adversarial Training algorithm using an AC-GAN to generate adversarial examples. By enhancing the AC-GAN architecture and comparing its performance to other models, it is shown that generative models are better suited for boosting adversarial security through adversarial training.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Analytical
Mohammed Mallik, Angesom Ataklity Tesfay, Benjamin Allaert, Redha Kassi, Esteban Egea-Lopez, Jose-Maria Molina-Garcia-Pardo, Joe Wiart, Davy P. Gaillot, Laurent Clavier
Summary: This study proposes a conditional generative adversarial network to accurately reconstruct the electromagnetic field exposure map in an outdoor urban environment. By learning the topology of the environment, the model is able to estimate the propagation characteristics of the electromagnetic field, and the results show that it outperforms traditional kriging methods in terms of accuracy.
Article
Medicine, General & Internal
Amal Al-Rasheed, Amel Ksibi, Manel Ayadi, Abdullah I. A. Alzahrani, Mohammed Zakariah, Nada Ali Hakami
Summary: Skin cancer is a severe disease that requires early detection and treatment. Manual diagnosis is time-consuming and prone to errors, hence automated diagnostic systems are needed. By fine-tuning deep learning models VGG16, ResNet50, and ResNet101 and utilizing data augmentation and generative adversarial networks, high accuracy in skin cancer classification can be achieved.
Article
Computer Science, Information Systems
Alaa Abu-Srhan, Mohammad A. M. Abushariah, Omar S. Al-Kadi
Summary: Conditional Generative Adversarial Network (cGAN) is modified by combining adversarial loss with non-adversarial loss functions to improve model performance in image-to-image translation tasks. The best combination of loss functions for image-to-image translation on the Facades dataset is WGAN adversarial loss with L1 and content non-adversarial loss functions. The model generates fine structure images and captures both high and low frequency details of translated images. The practicality of the proposed work is demonstrated through image in-painting and lesion segmentation.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Automation & Control Systems
Enjamamul Hoq, Osama Aljarrah, Jun Li, Jing Bi, Alfa Heryudono, Wenzhen Huang
Summary: This article explores different methods for predicting full stress fields in random heterogeneous materials, including model order reduction with classical machine learning and computer vision-based deep learning. The study finds that deep learning methods provide more accurate predictions with reduced errors compared to classical machine learning techniques.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Engineering, Industrial
Nathan Hertlein, Philip R. Buskohl, Andrew Gillman, Kumar Vemaganti, Sam Anand
Summary: This study introduces a deep learning-based framework that predicts near optimal designs by learning latent similarities between runs in a training set using a conditional generative adversarial network (cGAN). The cGAN predictions show good agreement with true optima, and for greater accuracy, predictions can be further improved by applying a small number of topology optimization iterations.
JOURNAL OF MANUFACTURING SYSTEMS
(2021)
Article
Engineering, Civil
Xu Hong, Liang Hu, Ahsan Kareem
Summary: This study proposes a model called CWGAN-GP for predicting tropical cyclone intensity evolution and assessing related risks. The model represents the nonlinear system of TC intensity evolution as a random variable conditioned on input variables characterizing the TC state and environment. It successfully replicates non-Gaussian probabilistic properties and shows potential in TC hazard assessment.
JOURNAL OF WIND ENGINEERING AND INDUSTRIAL AERODYNAMICS
(2023)
Article
Materials Science, Multidisciplinary
Soo Young Lee, Seokyeong Byeon, Hyoung Seop Kim, Hyungyu Jin, Seungchul Lee
Summary: Researchers utilize deep learning methods to predict phases of high-entropy alloys, achieving significant improvement in prediction accuracy and guidance for design through optimization, generation, and explanation of phase information.
MATERIALS & DESIGN
(2021)
Article
Computer Science, Interdisciplinary Applications
Eun-A Sim, Seunghye Lee, Jeongmin Oh, Jaehong Lee
Summary: This paper presents a novel combination of Generative Adversarial Networks (GANs) and Clustering Analysis (CA) for topology optimization, which generates new data and selects optimized data through clustering analysis. A Topology Optimization Validation Curve (TOVC) is successfully developed through the entire volume fraction of the structure, demonstrating the adaptability and efficiency of the proposed method for topology optimization.
ADVANCES IN ENGINEERING SOFTWARE
(2021)
Article
Thermodynamics
Yaning Wang, Zirui Wang, Wen Wang, Guocheng Tao, Weiqi Shen, Jiahuan Cui
Summary: A Conditional Generative Adversarial Network (CGAN) model was developed in this study to accurately predict the two-dimensional film cooling effectiveness based on the film cooling parameters. Mapping the characteristics of the trench hole and utilizing adversarial training improved the accuracy of the prediction.
INTERNATIONAL JOURNAL OF THERMAL SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Feixiang Liu, Yiru Dai
Summary: This paper conducts research in two aspects of data augmentation and model optimization, proposing a data generation model RVAE-CGAN and a product quality prediction model PPO-SVR. By combining the two models, the prediction effect of the product quality prediction model is significantly improved in the small sample data environment.
ADVANCED ENGINEERING INFORMATICS
(2023)
Article
Engineering, Mechanical
Mohammad Mahdi Behzadi, Horea T. Ilies
Summary: This paper proposes a design exploration framework for topology optimization that utilizes transfer learning and conditional generative adversarial networks (GANs) to significantly improve generalization ability and reduce data and computational resource requirements. Additionally, the inclusion of topological metrics in the loss function improves the connectivity of the predicted optimal topology.
JOURNAL OF MECHANICAL DESIGN
(2022)
Article
Computer Science, Interdisciplinary Applications
Qiyue Wang, Wu Xue, Xiaoke Zhang, Fang Jin, James Hahn
Summary: This paper introduces a method for predicting body composition using deep neural network and conditional generative adversarial network, achieving high accuracy. Experimental results demonstrate that the proposed method outperforms competitive methods significantly.
JOURNAL OF BIOMEDICAL INFORMATICS
(2021)