paper-with-me

홈 › Papers

On the Interplay of Priors and Overparametrization in Bayesian Neural Network Posteriors

2026-03-23 · Julius Kobialka, Emanuel Sommer, Chris Kolb, Juntae Kwon, Daniel Dold, David Rügamer arxiv

Bayesian neural network (BNN) posteriors are often considered impractical for inference, as symmetries fragment them, non-identifiabilities inflate dimensionality, and weight-space priors are seen as meaningless. In this work, we study how overparametrization and priors together reshape BNN posteriors and derive implications allowing us to better understand their interplay. We show that redundancy introduces three key phenomena that fundamentally reshape the posterior geometry: balancedness, weight reallocation on equal-probability manifolds, and prior conformity. We validate our findings through extensive experiments with posterior sampling budgets that far exceed those of earlier works, and demonstrate how overparametrization induces structured, prior-aligned weight posterior distributions.

📄 PDF Abstract BibTeX arXiv:2603.22030

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

The Problem of the Priors, or Posteriors?

2025-03-14 · Hanti Lin

The problem of the priors is well known: it concerns the challenge of identifying norms that govern one's prior credences. I argue that a key to addressing this problem lies in considering what I call the problem of the …

Philosophy

Monte Carlo guided Diffusion for Bayesian linear inverse problems

2023-08-15 · Gabriel Cardoso, Yazid Janati El Idrissi, Sylvain Le Corff, Eric Moulines

Ill-posed linear inverse problems arise frequently in various applications, from computational photography to medical imaging. A recent line of research exploits Bayesian inference with informative priors to handle the i…

Bayesian Inference

Gaussian processes for Bayesian inverse problems associated with linear partial differential equations

2023-07-17 · Tianming Bai, Aretha L. Teckentrup, Konstantinos C. Zygalakis

This work is concerned with the use of Gaussian surrogate models for Bayesian inverse problems associated with linear partial differential equations. A particular focus is on the regime where only a small amount of train…

Gaussian Processes

Meta-Learning Bayesian Neural Network Priors Based on PAC-Bayesian Theory

2021-01-01 · Jonas Rothfuss, Martin Josifoski, Andreas Krause

Bayesian deep learning is a promising approach towards improved uncertainty quantification and sample efficiency. Due to their complex parameter space, choosing informative priors for Bayesian Neural Networks (BNNs) is c…

Meta-LearningUncertainty QuantificationVariational Inference

Generalized Variational Inference: Three arguments for deriving new Posteriors

2019-04-03 · Jeremias Knoblauch, Jack Jewson, Theodoros Damoulas

We advocate an optimization-centric view on and introduce a novel generalization of Bayesian inference. Our inspiration is the representation of Bayes' rule as infinite-dimensional optimization problem (Csiszar, 1975; Do…

Bayesian InferenceGaussian ProcessesVariational Inference