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) are being widely applied for various real-worldapplications across domains due to their high performance (e.g., high accuracyon image classification). Nevertheless, a well-trained DNN after deploymentcould oftentimes raise errors during practical use in the operationalenvironment due to the mismatching between distributions of the trainingdataset and the potential unknown noise factors in the operational environment,e.g., weather, blur, noise etc. Hence, it poses a rather important problem forthe DNNs' real-world applications: how to repair the deployed DNNs forcorrecting the failure samples (i.e., incorrect prediction) under the deployedoperational environment while not harming their capability of handling normalor clean data. The number of failure samples we can collect in practice, causedby the noise factors in the operational environment, is often limited.Therefore, It is rather challenging how to repair more similar failures basedon the limited failure samples we can collect.  In this paper, we propose a style-guided data augmentation for repairing DNNin the operational environment. We propose a style transfer method to learn andintroduce the unknown failure patterns within the failure data into thetraining data via data augmentation. Moreover, we further propose theclustering-based failure data generation for much more effective style-guideddata augmentation. We conduct a large-scale evaluation with fifteen degradationfactors that may happen in the real world and compare with fourstate-of-the-art data augmentation methods and two DNN repairing methods,demonstrating that our method can significantly enhance the deployed DNNs onthe corrupted data in the operational environment, and with even betteraccuracy on clean datasets.

DeepRepair: Style-Guided Repairing for DNNs in the Real-world Operational Environment