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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep neural networks (DNNs) have achieved state-of-the-art performances inmany important domains, including medical diagnosis, security, and autonomousdriving. In these domains where safety is highly critical, an erroneousdecision can result in serious consequences. While a perfect predictionaccuracy is not always achievable, recent work on Bayesian deep networks showsthat it is possible to know when DNNs are more likely to make mistakes. Knowingwhat DNNs do not know is desirable to increase the safety of deep learningtechnology in sensitive applications. Bayesian neural networks attempt toaddress this challenge. However, traditional approaches are computationallyintractable and do not scale well to large, complex neural networkarchitectures. In this paper, we develop a theoretical framework to approximateBayesian inference for DNNs by imposing a Bernoulli distribution on the modelweights. This method, called MC-DropConnect, gives us a tool to represent themodel uncertainty with little change in the overall model structure orcomputational cost. We extensively validate the proposed algorithm on multiplenetwork architectures and datasets for classification and semantic segmentationtasks. We also propose new metrics to quantify the uncertainty estimates. Thisenables an objective comparison between MC-DropConnect and prior approaches.Our empirical results demonstrate that the proposed framework yieldssignificant improvement in both prediction accuracy and uncertainty estimationquality compared to the state of the art.

DropConnect Is Effective in Modeling Uncertainty of Bayesian Deep Networks