EUVIMEDEUVIMED entdecken
Crossref

Implicit generative prior for Bayesian neural networks

Yijia Liu, Xiao Wang

The Annals of Applied Statistics · 01.12.2024 · 18 (4) · 2840-2862

DOI: 10.1214/24-aoas1908 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Annals of Applied Statistics, erschienen am 01.12.2024. Verfasst von Yijia Liu und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 61 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Predictive uncertainty quantification is crucial for reliable decision-making in various applied domains. Bayesian neural networks offer a powerful framework for this task. However, defining meaningful priors and ensuring computational efficiency remain significant challenges, especially for complex real-world applications. This paper addresses these challenges by proposing a novel neural adaptive empirical Bayes (NA-EB) framework. NA-EB leverages a class of implicit generative priors derived from low-dimensional distributions. This allows for efficient handling of complex data structures and effective capture of underlying relationships in real-world datasets. The proposed NA-EB framework combines variational inference with a gradient ascent algorithm. This enables simultaneous hyperparameter selection and approximation of the posterior distribution, leading to improved computational efficiency. We establish the theoretical foundation of the framework through posterior and classification consistency. We demonstrate the practical applications of our framework through extensive evaluations on a variety of tasks, including the two-spiral problem, regression, 10 UCI datasets, and image classification tasks on both MNIST and CIFAR-10 datasets. The results of our experiments highlight the superiority of our proposed framework over existing methods, such as sparse variational Bayesian and generative models, in terms of prediction accuracy and uncertainty quantification.

Zitieren

APA

Liu, Y., & Wang, X. (2024). Implicit generative prior for Bayesian neural networks. The Annals of Applied Statistics, 18(4), 2840-2862. https://doi.org/10.1214/24-aoas1908

Vancouver

Liu Y, Wang X. Implicit generative prior for Bayesian neural networks. The Annals of Applied Statistics. 2024;18(4):2840-2862. doi:10.1214/24-aoas1908

BibTeX
@article{Liu2024,
  author = {Liu, Yijia and Wang, Xiao},
  title = {Implicit generative prior for Bayesian neural networks},
  journal = {The Annals of Applied Statistics},
  year = {2024},
  volume = {18},
  number = {4},
  pages = {2840--2862},
  doi = {10.1214/24-aoas1908}
}
Kostenlos starten →