Generative Adversarial Network
Generative Adversarial Networks (GANs) are a class of deep learning models designed to generate new data instances that resemble a training dataset. Current research focuses on improving GAN training stability, enhancing the quality and diversity of generated data, and applying GANs to diverse fields like medical imaging, drug discovery, and time series analysis, often incorporating techniques like contrastive learning and disentangled representation learning to improve model performance and interpretability. The ability of GANs to synthesize realistic data addresses critical limitations in data availability and annotation costs across numerous scientific disciplines and practical applications, leading to advancements in areas ranging from medical diagnosis to robotic control.
Papers
Unaligned 2D to 3D Translation with Conditional Vector-Quantized Code Diffusion using Transformers
Abril Corona-Figueroa, Sam Bond-Taylor, Neelanjan Bhowmik, Yona Falinie A. Gaus, Toby P. Breckon, Hubert P. H. Shum, Chris G. Willcocks
A Bayesian Non-parametric Approach to Generative Models: Integrating Variational Autoencoder and Generative Adversarial Networks using Wasserstein and Maximum Mean Discrepancy
Forough Fazeli-Asl, Michael Minyi Zhang
PFL-GAN: When Client Heterogeneity Meets Generative Models in Personalized Federated Learning
Achintha Wijesinghe, Songyang Zhang, Zhi Ding
A Systematic Study on Quantifying Bias in GAN-Augmented Data
Denis Liu
CoC-GAN: Employing Context Cluster for Unveiling a New Pathway in Image Generation
Zihao Wang, Yiming Huang, Ziyu Zhou
Deep Generative Modeling-based Data Augmentation with Demonstration using the BFBT Benchmark Void Fraction Datasets
Farah Alsafadi, Xu Wu
Physics-guided training of GAN to improve accuracy in airfoil design synthesis
Kazunari Wada, Katsuyuki Suzuki, Kazuo Yonekura
EGANS: Evolutionary Generative Adversarial Network Search for Zero-Shot Learning
Shiming Chen, Shihuang Chen, Wenjin Hou, Weiping Ding, Xinge You