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

Biases can filter into AI technology without our knowledge. Oftentimes,seminal deep learning networks champion increased accuracy above all else. Inthis paper, we attempt to alleviate biases encountered by semantic segmentationmodels in urban driving scenes, via an iteratively trained unlearningalgorithm. Convolutional neural networks have been shown to rely on colour andtexture rather than geometry. This raises issues when safety-criticalapplications, such as self-driving cars, encounter images with covariate shiftat test time - induced by variations such as lighting changes or seasonality.Conceptual proof of bias unlearning has been shown on simple datasets such asMNIST. However, the strategy has never been applied to the safety-criticaldomain of pixel-wise semantic segmentation of highly variable training data -such as urban scenes. Trained models for both the baseline and bias unlearningscheme have been tested for performance on colour-manipulated validation setsshowing a disparity of up to 85.50% in mIoU from the original RGB images -confirming segmentation networks strongly depend on the colour information inthe training data to make their classification. The bias unlearning schemeshows improvements of handling this covariate shift of up to 61% in the bestobserved case - and performs consistently better at classifying the "human" and"vehicle" classes compared to the baseline model.

"Just Drive": Colour Bias Mitigation for Semantic Segmentation in the Context of Urban Driving