Towards Requirements Specification for Machine-learned Perception Based on Human Performance
Author ORCID Identifier
Publication Title
Proceedings - 7th International Workshop on Artificial Intelligence and Requirements Engineering, AIRE 2020
ISBN
9781728183527
Document Type
Conference Proceeding
Abstract
The application of machine learning (ML) based perception algorithms in safety-critical systems such as autonomous vehicles have raised major safety concerns due to the apparent risks to human lives. Yet assuring the safety of such systems is a challenging task, in a large part because ML components (MLCs) rarely have clearly specified requirements. Instead, they learn their intended tasks from the training data. One of the most well-studied properties that ensure the safety of MLCs is the robustness against small changes in images. But the range of changes considered small has not been systematically defined. In this paper, we propose an approach for specifying and testing requirements for robustness based on human perception. With this approach, the MLCs are required to be robust to changes that fall within the range defined based on human perception performance studies. We demonstrate the approach on a state-of-the-art object detector.
First Page
48
Last Page
51
Publication Date
9-1-2020
DOI
10.1109/AIRE51212.2020.00014
Recommended Citation
Hu, Boyue C.; Salay, Rick; Czarnecki, Krzysztof; Rahimi, Mona; Selim, Gehan; and Chechik, Marsha, "Towards Requirements Specification for Machine-learned Perception Based on Human Performance" (2020). NIU Bibliography. 652.
https://huskiecommons.lib.niu.edu/niubib/652
Department
Department of Computer Science
