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

The quality of Artificial Intelligence (AI) algorithms is of significantimportance for confidently adopting algorithms in various applications such ascybersecurity, healthcare, and autonomous driving. This work presents aprincipled framework of using a design-of-experimental approach tosystematically evaluate the quality of AI algorithms, named as Do-AIQ.Specifically, we focus on investigating the quality of the AI mislabel dataalgorithm against data poisoning. The performance of AI algorithms is affectedby hyperparameters in the algorithm and data quality, particularly, datamislabeling, class imbalance, and data types. To evaluate the quality of the AIalgorithms and obtain a trustworthy assessment on the quality of thealgorithms, we establish a design-of-experiment framework to construct anefficient space-filling design in a high-dimensional constraint space anddevelop an effective surrogate model using additive Gaussian process to enablethe emulation of the quality of AI algorithms. Both theoretical and numericalstudies are conducted to justify the merits of the proposed framework. Theproposed framework can set an exemplar for AI algorithm to enhance the AIassurance of robustness, reproducibility, and transparency.

Do-AIQ: A Design-of-Experiment Approach to Quality Evaluation of AI Mislabel Detection Algorithm