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

A critical step to building trustworthy deep neural networks is trustquantification, where we ask the question: How much can we trust a deep neuralnetwork? In this study, we take a step towards simple, interpretable metricsfor trust quantification by introducing a suite of metrics for assessing theoverall trustworthiness of deep neural networks based on their behaviour whenanswering a set of questions. We conduct a thought experiment and explore twokey questions about trust in relation to confidence: 1) How much trust do wehave in actors who give wrong answers with great confidence? and 2) How muchtrust do we have in actors who give right answers hesitantly? Based on insightsgained, we introduce the concept of question-answer trust to quantifytrustworthiness of an individual answer based on confident behaviour undercorrect and incorrect answer scenarios, and the concept of trust density tocharacterize the distribution of overall trust for an individual answerscenario. We further introduce the concept of trust spectrum for representingoverall trust with respect to the spectrum of possible answer scenarios acrosscorrectly and incorrectly answered questions. Finally, we introduceNetTrustScore, a scalar metric summarizing overall trustworthiness. The suiteof metrics aligns with past social psychology studies that study therelationship between trust and confidence. Leveraging these metrics, wequantify the trustworthiness of several well-known deep neural networkarchitectures for image recognition to get a deeper understanding of wheretrust breaks down. The proposed metrics are by no means perfect, but the hopeis to push the conversation towards better metrics to help guide practitionersand regulators in producing, deploying, and certifying deep learning solutionsthat can be trusted to operate in real-world, mission-critical scenarios.

How Much Can We Really Trust You? Towards Simple, Interpretable Trust Quantification Metrics for Deep Neural Networks