Fusion framework and multimodality for the Laplacian approximation of Bayesian neural networks

TL;DR

This paper introduces a fusion framework using multimodal Laplacian approximations of Bayesian neural networks to improve robustness and uncertainty calibration in image classification tasks.

Contribution

It extends Laplacian approximation of Bayesian NNs to represent multimodal distributions, enhancing uncertainty estimation and fusion strategies.

Findings

Improved calibration of uncertainty estimates.

Enhanced robustness through multimodal fusion.

Validated on MNIST, CIFAR-10, and camera trap data.

Abstract

This paper considers the problem of sequential fusion of predictions from neural networks (NN) and fusion of predictions from multiple NN. This fusion strategy increases the robustness, i.e., reduces the impact of one incorrect classification and detection of outliers the \nn has not seen during training. This paper uses Laplacian approximation of Bayesian NNs (BNNs) to quantify the uncertainty necessary for fusion. Here, an extension is proposed such that the prediction of the NN can be represented by multimodal distributions. Regarding calibration of the estimated uncertainty in the prediction, the performance is significantly improved by having the flexibility to represent a multimodal distribution. Two class classical image classification tasks, i.e., MNIST and CFAR10, and image sequences from camera traps of carnivores in Swedish forests have been used to demonstrate the fusion strategies and proposed extension to the Laplacian approximation.