Calibration of deep probabilistic models with decoupled bayesian neural networks

Deep Neural Networks (DNNs) have achieved state-of-the-art accuracy performance in many tasks. However, recent works have pointed out that the outputs provided by these models are not well-calibrated, seriously limiting their use in critical decision scenarios. In this work, we propose to use a deco...

Descripción completa

Detalles Bibliográficos
Autores: Maroñas Molano, Juan, Paredes, Roberto, Ramos Castro, Daniel
Tipo de recurso: artículo
Fecha de publicación:2020
País:España
Institución:Universidad Autónoma de Madrid
Repositorio:Biblos-e Archivo. Repositorio Institucional de la UAM
Idioma:inglés
OAI Identifier:oai:repositorio.uam.es:10486/703083
Acceso en línea:http://hdl.handle.net/10486/703083
https://dx.doi.org/10.1016/j.neucom.2020.04.103
Access Level:acceso abierto
Palabra clave:Bayesian modelling
Bayesian neural networks
Calibration
Image classification
Telecomunicaciones
Descripción
Sumario:Deep Neural Networks (DNNs) have achieved state-of-the-art accuracy performance in many tasks. However, recent works have pointed out that the outputs provided by these models are not well-calibrated, seriously limiting their use in critical decision scenarios. In this work, we propose to use a decoupled Bayesian stage, implemented with a Bayesian Neural Network (BNN), to map the uncalibrated probabilities provided by a DNN to calibrated ones, consistently improving calibration. Our results evidence that incorporating uncertainty provides more reliable probabilistic models, a critical condition for achieving good calibration. We report a generous collection of experimental results using high-accuracy DNNs in standardized image classification benchmarks, showing the good performance, flexibility and robust behaviour of our approach with respect to several state-of-the-art calibration methods. Code for reproducibility is provided