Saved in:
Bibliographic Details
Main Authors: Coscia, Dario, Welling, Max, Demo, Nicola, Rozza, Gianluigi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2501.18665
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911065370001408
author Coscia, Dario
Welling, Max
Demo, Nicola
Rozza, Gianluigi
author_facet Coscia, Dario
Welling, Max
Demo, Nicola
Rozza, Gianluigi
contents Autoregressive and recurrent networks have achieved remarkable progress across various fields, from weather forecasting to molecular generation and Large Language Models. Despite their strong predictive capabilities, these models lack a rigorous framework for addressing uncertainty, which is key in scientific applications such as PDE solving, molecular generation and Machine Learning Force Fields. To address this shortcoming we present BARNN: a variational Bayesian Autoregressive and Recurrent Neural Network. BARNNs aim to provide a principled way to turn any autoregressive or recurrent model into its Bayesian version. BARNN is based on the variational dropout method, allowing to apply it to large recurrent neural networks as well. We also introduce a temporal version of the "Variational Mixtures of Posteriors" prior (tVAMP-prior) to make Bayesian inference efficient and well-calibrated. Extensive experiments on PDE modelling and molecular generation demonstrate that BARNN not only achieves comparable or superior accuracy compared to existing methods, but also excels in uncertainty quantification and modelling long-range dependencies.
format Preprint
id arxiv_https___arxiv_org_abs_2501_18665
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BARNN: A Bayesian Autoregressive and Recurrent Neural Network
Coscia, Dario
Welling, Max
Demo, Nicola
Rozza, Gianluigi
Machine Learning
Artificial Intelligence
Autoregressive and recurrent networks have achieved remarkable progress across various fields, from weather forecasting to molecular generation and Large Language Models. Despite their strong predictive capabilities, these models lack a rigorous framework for addressing uncertainty, which is key in scientific applications such as PDE solving, molecular generation and Machine Learning Force Fields. To address this shortcoming we present BARNN: a variational Bayesian Autoregressive and Recurrent Neural Network. BARNNs aim to provide a principled way to turn any autoregressive or recurrent model into its Bayesian version. BARNN is based on the variational dropout method, allowing to apply it to large recurrent neural networks as well. We also introduce a temporal version of the "Variational Mixtures of Posteriors" prior (tVAMP-prior) to make Bayesian inference efficient and well-calibrated. Extensive experiments on PDE modelling and molecular generation demonstrate that BARNN not only achieves comparable or superior accuracy compared to existing methods, but also excels in uncertainty quantification and modelling long-range dependencies.
title BARNN: A Bayesian Autoregressive and Recurrent Neural Network
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2501.18665