File Download

There are no files associated with this item.

Supplementary

Conference Paper: Humans vs. AI in Detecting Vehicles and Humans in Driving Scenarios

TitleHumans vs. AI in Detecting Vehicles and Humans in Driving Scenarios
Authors
Issue Date26-Jul-2023
Abstract

To inform Explainable AI (XAI) design for updating users’ beliefs about AI based on their mental models, we examined the similarities and differences between humans and AI in object detection in driving scenarios. In humans, individuals differed in adopting focused or explorative attention strategies, with better performance associated with the focused strategy. AI (Yolo-v5s) had higher similarity in attended features to the focused than the explorative strategy in humans, and achieved human-expert-level performance in vehicle detection even in difficult cases such as occlusion and degradation. In contrast, it performed much poorer than humans in detecting humans with low attended feature similarity due to humans’ attention bias for stimuli with evolutionary significance. Also, higher similarity to humans’ attended features was associated with better AI performance, suggesting that human attention may be used for guiding AI design. These findings have significant implications for both AI and XAI designs.


Persistent Identifierhttp://hdl.handle.net/10722/337706

 

DC FieldValueLanguage
dc.contributor.authorYang, A-
dc.contributor.authorLiu, G-
dc.contributor.authorChen, Y-
dc.contributor.authorQi, R-
dc.contributor.authorZhang, J-
dc.contributor.authorHsiao, J H-
dc.date.accessioned2024-03-11T10:23:14Z-
dc.date.available2024-03-11T10:23:14Z-
dc.date.issued2023-07-26-
dc.identifier.urihttp://hdl.handle.net/10722/337706-
dc.description.abstract<p>To inform Explainable AI (XAI) design for updating users’ beliefs about AI based on their mental models, we examined the similarities and differences between humans and AI in object detection in driving scenarios. In humans, individuals differed in adopting focused or explorative attention strategies, with better performance associated with the focused strategy. AI (Yolo-v5s) had higher similarity in attended features to the focused than the explorative strategy in humans, and achieved human-expert-level performance in vehicle detection even in difficult cases such as occlusion and degradation. In contrast, it performed much poorer than humans in detecting humans with low attended feature similarity due to humans’ attention bias for stimuli with evolutionary significance. Also, higher similarity to humans’ attended features was associated with better AI performance, suggesting that human attention may be used for guiding AI design. These findings have significant implications for both AI and XAI designs.</p>-
dc.languageeng-
dc.relation.ispartof44th Annual Meeting of the Cognitive Science Society (26/07/2023-29/07/2023, Sydney)-
dc.titleHumans vs. AI in Detecting Vehicles and Humans in Driving Scenarios-
dc.typeConference_Paper-
dc.identifier.issue45-
dc.identifier.spage1832-
dc.identifier.epage1839-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats