Inducing Metrizability of Generative Adversarial Networks with Discriminative Normalized Linear Layer
Generative adversarial networks (GANs) can be optimized to make the generator distribution close to the target distribution by satisfying metrizable conditions on the discriminator, including direction optimality, separability, and injectivity.
Improved Techniques for Enhancing Likelihood Estimation in Diffusion Ordinary Differential Equation Models
The authors propose several improved techniques, including both the evaluation perspective and training perspective, to allow the likelihood estimation by diffusion ODEs to outperform the existing state-of-the-art likelihood estimators.
Stability of Iterative Retraining of Generative Models on Their Own Data
Iterative retraining of generative models on a mix of real and synthetic data can be stable, provided the initial generative model is sufficiently well-trained and the proportion of real data is large enough.
Scalable Wasserstein Gradient Flow for Generative Modeling through Unbalanced Optimal Transport
새로운 Scalable WGF 기반 생성 모델인 Semi-dual JKO를 소개합니다.
Optimal-Transport-based Adversarial Networks: Analysis and Improvement
Optimal Transport theory enhances stability and performance in generative modeling.
Analyzing Simplified Diffusion Schrödinger Bridge for Enhanced Generative Modeling
Introducing a simplified approach to Diffusion Schrödinger Bridge for improved generative modeling.
Optimal-Transport-based Adversarial Networks Analysis and Improvement
OT-based GANs benefit from strictly convex functions and cost functions to enhance stability and prevent mode collapse.
© 2024 by Linnk AI