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

Despite the recent progress in deep neural networks (DNNs), it remainschallenging to explain the predictions made by DNNs. Existing explanationmethods for DNNs mainly focus on post-hoc explanations where anotherexplanatory model is employed to provide explanations. The fact that post-hocmethods can fail to reveal the actual original reasoning process of DNNs raisesthe need to build DNNs with built-in interpretability. Motivated by this, manyself-explaining neural networks have been proposed to generate not onlyaccurate predictions but also clear and intuitive insights into why aparticular decision was made. However, existing self-explaining networks arelimited in providing distribution-free uncertainty quantification for the twosimultaneously generated prediction outcomes (i.e., a sample's final predictionand its corresponding explanations for interpreting that prediction).Importantly, they also fail to establish a connection between the confidencevalues assigned to the generated explanations in the interpretation layer andthose allocated to the final predictions in the ultimate prediction layer. Totackle the aforementioned challenges, in this paper, we design a noveluncertainty modeling framework for self-explaining networks, which not onlydemonstrates strong distribution-free uncertainty modeling performance for thegenerated explanations in the interpretation layer but also excels in producingefficient and effective prediction sets for the final predictions based on theinformative high-level basis explanations. We perform the theoretical analysisfor the proposed framework. Extensive experimental evaluation demonstrates theeffectiveness of the proposed uncertainty framework.

Towards Modeling Uncertainties of Self-explaining Neural Networks via Conformal Prediction