Improved Surprise Adequacy Tools for Corner Case Data Description and Detection
| Autor: | Yutaka Oiwa, Tinghui Ouyang, Yoshinao Isobe, Vicent Sanz Marco, Hideki Asoh, Yoshiki Seo |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Technology
Computer science QH301-705.5 020209 energy media_common.quotation_subject QC1-999 AI quality testing 02 engineering and technology computer.software_genre Data description Corner case Factor (programming language) surprise adequacy 0202 electrical engineering electronic engineering information engineering General Materials Science Biology (General) Instrumentation QD1-999 media_common computer.programming_language Fluid Flow and Transfer Processes business.industry Process Chemistry and Technology Physics General Engineering Construct (python library) 021001 nanoscience & nanotechnology Engineering (General). Civil engineering (General) corner case data detection Computer Science Applications Surprise Improved performance Chemistry Data mining TA1-2040 0210 nano-technology business modified distanced-based SA Quality assurance computer MNIST database |
| Zdroj: | Applied Sciences, Vol 11, Iss 6826, p 6826 (2021) Applied Sciences Volume 11 Issue 15 |
| ISSN: | 2076-3417 |
| Popis: | Facing the increasing quantity of AI models applications, especially in life- and property-related fields, it is crucial for designers to construct safety- and security-critical systems. As a major factor affecting the safety of AI models, corner case data and its related description/detection techniques are important in the AI design phase and quality assurance. In this paper, inspired by surprise adequacy (SA), a tool having advantages on capture data behaviors, we developed three modified versions of distance-based-SA (DSA) for detecting corner cases in classification problems. Through the experiment analysis on MNIST, CIFAR, and industrial example data, the feasibility and usefulness of the proposed tools on corner case data detection are verified. Moreover, Qualitative and quantitative experiments validated that the developed DSA tools can achieve improved performance in describing corner cases’ behaviors. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|