JpGU-AGU Joint Meeting 2020

Presentation information

[E] Oral

A (Atmospheric and Hydrospheric Sciences ) » A-CG Complex & General

[A-CG50] Earth & Environmental Sciences and Artificial Intelligence

convener:Tomohiko Tomita(Faculty of Advanced Science and Technology, Kumamoto University), Shigeki Hosoda(Japan Marine-Earth Science and Technology), Ken-ichi Fukui(Osaka University), Satoshi Ono(Kagoshima University)

[ACG50-01] What is "Explainable AI"?

★Invited Papers

*Satoshi Hara1 (1.Osaka University)

Keywords:Artificial Inteligence, Explainable AI, Machine Learning

In this talk, I will present a brief overview of "Explainable AI". The recent developments of AI technologies enable computers to perform comparably or even better than the human on some tasks such as image recognition. However, the "black-box" nature of AI is considered as a crucial shortcoming of the current AI. That is, current AI only outputs its prediction or recognition results but it does not explain why and how such outputs are obtained in a "human interpretable" manner. To alleviate the "black-box" nature, several efforts have been made to AI to be "explainable" in the last few years. I will introduce these recent developments on "Explainable AI".