English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Neural network ensembles, such as Bayesian neural networks (BNNs), have shownsuccess in the areas of uncertainty estimation and robustness. However, acrucial challenge prohibits their use in practice. BNNs require a large numberof predictions to produce reliable results, leading to a significant increasein computational cost. To alleviate this issue, we propose spatial smoothing, amethod that spatially ensembles neighboring feature map points of convolutionalneural networks. By simply adding a few blur layers to the models, weempirically show that spatial smoothing improves accuracy, uncertaintyestimation, and robustness of BNNs across a whole range of ensemble sizes. Inparticular, BNNs incorporating spatial smoothing achieve high predictiveperformance merely with a handful of ensembles. Moreover, this method also canbe applied to canonical deterministic neural networks to improve theperformances. A number of evidences suggest that the improvements can beattributed to the stabilized feature maps and the smoothing of the losslandscape. In addition, we provide a fundamental explanation for prior works -namely, global average pooling, pre-activation, and ReLU6 - by addressing themas special cases of spatial smoothing. These not only enhance accuracy, butalso improve uncertainty estimation and robustness by making the loss landscapesmoother in the same manner as spatial smoothing. The code is available athttps://github.com/xxxnell/spatial-smoothing.

Blurs Behave Like Ensembles: Spatial Smoothings to Improve Accuracy, Uncertainty, and Robustness