TY - GEN
T1 - 'Why did my AI agent lose?'
T2 - 2021 IEEE Visualization Conference, VIS 2021
AU - Tabatabai, Delyar
AU - Ruangrotsakun, Anita
AU - Irvine, Jed
AU - Dodge, Jonathan
AU - Shureih, Zeyad
AU - Lam, Kin Ho
AU - Burnett, Margaret
AU - Fern, Alan
AU - Kahng, Minsuk
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - How can we help domain-knowledgeable users who do not have expertise in AI analyze why an AI agent failed? Our research team previously developed a new structured process for such users to assess AI, called After-Action Review for AI (AAR/AI), consisting of a series of steps a human takes to assess an AI agent and formalize their understanding. In this paper, we investigate how the AAR/AI process can scale up to support reinforcement learning (RL) agents that operate in complex environments. We augment the AAR/AI process to be performed at three levels - episode-level, decision-level, and explanation-level - and integrate it into our redesigned visual analytics interface. We illustrate our approach through a usage scenario of analyzing why a RL agent lost in a complex real-time strategy game built with the StarCraft 2 engine. We believe integrating structured processes like AAR/AI into visualization tools can help visualization play a more critical role in AI interpretability.
AB - How can we help domain-knowledgeable users who do not have expertise in AI analyze why an AI agent failed? Our research team previously developed a new structured process for such users to assess AI, called After-Action Review for AI (AAR/AI), consisting of a series of steps a human takes to assess an AI agent and formalize their understanding. In this paper, we investigate how the AAR/AI process can scale up to support reinforcement learning (RL) agents that operate in complex environments. We augment the AAR/AI process to be performed at three levels - episode-level, decision-level, and explanation-level - and integrate it into our redesigned visual analytics interface. We illustrate our approach through a usage scenario of analyzing why a RL agent lost in a complex real-time strategy game built with the StarCraft 2 engine. We believe integrating structured processes like AAR/AI into visualization tools can help visualization play a more critical role in AI interpretability.
UR - http://www.scopus.com/inward/record.url?scp=85123777036&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85123777036&partnerID=8YFLogxK
U2 - 10.1109/VIS49827.2021.9623268
DO - 10.1109/VIS49827.2021.9623268
M3 - Conference contribution
AN - SCOPUS:85123777036
T3 - Proceedings - 2021 IEEE Visualization Conference - Short Papers, VIS 2021
SP - 16
EP - 20
BT - Proceedings - 2021 IEEE Visualization Conference - Short Papers, VIS 2021
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 24 October 2021 through 29 October 2021
ER -