Bayesian Neural Networks (BNNs) have long been considered an ideal, yet unscalable solution for improving the robustness and the predictive uncertainty of deep neural networks. While they could capture more accurately the posterior distribution of the network parameters, most BNN approaches are either limited to small networks or rely on constraining assumptions, e.g., parameter independence. These drawbacks have enabled prominence of simple, but computationally heavy approaches such as Deep Ensembles, whose training and testing costs increase linearly with the number of networks. In this work we aim for efficient deep BNNs amenable to complex computer vision architectures, e.g., ResNet-50 DeepLabv3+, and tasks, e.g., semantic segmentation and image classification, with fewer assumptions on the parameters. We achieve this by leveraging variational autoencoders (VAEs) to learn the interaction and the latent distribution of the parameters at each network layer. Our approach, called Latent-Posterior BNN (LP-BNN), is compatible with the recent BatchEnsemble method, leading to highly efficient (in terms of computation and memory during both training and testing) ensembles. LP-BNNs attain competitive results across multiple metrics in several challenging benchmarks for image classification, semantic segmentation, and out-of-distribution detection.

Deep neural networks (DNNs) are powerful learning models, yet their results are not always reliable. This drawback results from the fact that modern DNNs are usually overconfident, and consequently their epistemic uncertainty cannot be straightforwardly characterized. In this work, we propose a new technique to quantify easily the epistemic uncertainty of data. This method consists in mixing the predictions of an ensemble of DNNs trained to classify One class versus All the other classes (OVA) with predictions from a standard DNN trained to perform All versus All (AVA) classification. First of all, the adjustment provided by the AVA DNN to the score of the base classifiers allows for a more fine-grained inter-class separation. Moreover, the two types of classifiers enforce mutually their detection of out-of-distribution (OOD) samples, circumventing entirely the requirement of using such samples during training. The additional cost involved by the construction of the ensemble is offset by the ease of use of our proposed strategy and by its enhanced generalization potential, as it does not bind its performance in a given context to specific OOD datasets. The extensive experiments confirm the wide applicability of our approach, and our method achieves state of the art performance in quantifying OOD data across multiple datasets and architectures while requiring little hyper-parameter tuning.

This article deals with the issue of reducing the spectral dimension of a hyperspectral image using principal component analysis (PCA). To perform this dimensionality reduction, we propose the addition of spatial information in order to improve the features that are extracted. Several approaches proposed to add spatial information are discussed in this article. They are based on mathematical morphology operators. These morphological operators are the area opening/closing, granulometries and grey-scale distance function. We name the proposed family of techniques the Morphological Principal Component Analysis (MorphPCA). Present approaches provide new feature spaces able to jointly handle the spatial and spectral information of hyperspectral images. They are computationally simple since the key element is the computation of an empirical covariance matrix which integrates simultaneously both spatial and spectral information. The performance of the different feature spaces is assessed for different tasks in order to prove their practical interest.