paper-with-me

Papers

Reparameterization invariance in approximate Bayesian inference

2024-06-05 · Hrittik Roy, Marco Miani, Carl Henrik Ek, Philipp Hennig, Marvin Pförtner, Lukas Tatzel, Søren Hauberg

Current approximate posteriors in Bayesian neural networks (BNNs) exhibit a crucial limitation: they fail to maintain invariance under reparameterization, i.e. BNNs assign different posterior densities to different parametrizations of identical functions. This creates a fundamental flaw in the application of Bayesian principles as it breaks the correspondence between uncertainty over the parameters with uncertainty over the parametrized function. In this paper, we investigate this issue in the context of the increasingly popular linearized Laplace approximation. Specifically, it has been observed that linearized predictives alleviate the common underfitting problems of the Laplace approximation. We develop a new geometric view of reparametrizations from which we explain the success of linearization. Moreover, we demonstrate that these reparameterization invariance properties can be extended to the original neural network predictive using a Riemannian diffusion process giving a straightforward algorithm for approximate posterior sampling, which empirically improves posterior fit.

📄 PDF Abstract BibTeX arXiv:2406.03334

Code (0)

등록된 구현이 없습니다.

Tasks

Bayesian Inference

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

Reparameterization Gradients through Acceptance-Rejection Sampling Algorithms

2016-10-18 · Christian A. Naesseth, Francisco J. R. Ruiz, Scott W. Linderman, David M. Blei

Variational inference using the reparameterization trick has enabled large-scale approximate Bayesian inference in complex probabilistic models, leveraging stochastic optimization to sidestep intractable expectations. Th…

Bayesian InferenceStochastic OptimizationVariational Inference

On the detrimental effect of invariances in the likelihood for variational inference

2022-09-15 · Richard Kurle, Ralf Herbrich, Tim Januschowski, Yuyang Wang 외

Variational Bayesian posterior inference often requires simplifying approximations such as mean-field parametrisation to ensure tractability. However, prior work has associated the variational mean-field approximation fo…

Variational Inference

A Coordinate-Free Construction of Scalable Natural Gradient

2018-08-30 · Kevin Luk, Roger Grosse

Most neural networks are trained using first-order optimization methods, which are sensitive to the parameterization of the model. Natural gradient descent is invariant to smooth reparameterizations because it is defined…

Pathologies in priors and inference for Bayesian transformers

2021-10-08 · NeurIPS Workshop ICBINB 2021 12 · Tristan Cinquin, Alexander Immer, Max Horn, Vincent Fortuin

In recent years, the transformer has established itself as a workhorse in many applications ranging from natural language processing to reinforcement learning. Similarly, Bayesian deep learning has become the gold-standa…

Bayesian InferenceVariational Inference

A Reparameterization-Invariant Flatness Measure for Deep Neural Networks

2019-11-29 · Henning Petzka, Linara Adilova, Michael Kamp, Cristian Sminchisescu

The performance of deep neural networks is often attributed to their automated, task-related feature construction. It remains an open question, though, why this leads to solutions with good generalization, even in cases …

Open-Ended Question Answering