Repository logo
 

Rethink reporting of evaluation results in AI.

Accepted version
Peer-reviewed

Type

Article

Change log

Authors

Burnell, Ryan 
Schellaert, Wout 
Ullman, Tomer D 
Martinez-Plumed, Fernando 

Abstract

Aggregate metrics and lack of access to results limit understanding.

Description

Keywords

46 Information and Computing Sciences, 4608 Human-Centred Computing, 3 Good Health and Well Being

Journal Title

Science

Conference Name

Journal ISSN

0036-8075
1095-9203

Volume Title

380

Publisher

American Association for the Advancement of Science (AAAS)
Sponsorship
Remove (HR00112120007)