Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
SIS, ITE-BCT |
2021-10-08 10:00 |
Online |
Online |
[Tutorial Lecture]
The Past and The Future of Explainable AI Techniques Yoshitaka Kameya (Meijo Univ.) SIS2021-17 |
Machine learning models of high predictive performance, such as deep neural networks and ensemble models, now play a cen... [more] |
SIS2021-17 pp.36-41 |
TL |
2021-09-18 16:05 |
Online |
Online |
[Keynote Address]
Neural Network as an Explainable Human
-- A New Approach to Contrastive Studies -- Yugo Murawaki (Kyoto Univ.) TL2021-16 |
In this talk, I argue that techniques developed in the field of explainable AI (XAI) have potential applications in comp... [more] |
TL2021-16 pp.23-27 |
MVE, IMQ, IE, CQ (Joint) [detail] |
2021-03-03 14:40 |
Online |
Online |
Semantic and Quantitative Explanation for Networks using Feature Interaction Bohui Xia, Xueting Wang, Toshihiko Yamasaki (The Univ. of Tokyo) IMQ2020-35 IE2020-75 MVE2020-67 |
Deep learning-based methods have shown remarkable performances in many tasks. However, it is hard for us to interpret th... [more] |
IMQ2020-35 IE2020-75 MVE2020-67 pp.127-132 |
PRMU |
2020-12-18 17:00 |
Online |
Online |
An evaluation method of area detection AI based on contribution pattern variation with noise addition Yasuhide Mori, Naofumi Hama, Masashi Egi (Hitachi) PRMU2020-67 |
The processing of image recognition AI using machine learning is generally black-boxed, and grasping the operating chara... [more] |
PRMU2020-67 pp.166-171 |
MI, IE, SIP, BioX, ITE-IST, ITE-ME [detail] |
2020-05-29 14:30 |
Online |
Online |
A method for analyze causes of deterioration of predict quality when Deep Learning is applied to instance segmentation Tomonori Kubota, Takanori Nakao, Masafumi Katoh, Eiji Yoshida, Hidenobu Miyoshi (Fujitsu Lab.) SIP2020-14 BioX2020-14 IE2020-14 MI2020-14 |
In this paper, we propose a method to analyze the cause of deterioration of prediction accuracy in instance segmentation... [more] |
SIP2020-14 BioX2020-14 IE2020-14 MI2020-14 pp.67-72 |
IE, IMQ, MVE, CQ (Joint) [detail] |
2020-03-06 14:50 |
Fukuoka |
Kyushu Institute of Technology (Cancelled but technical report was issued) |
A high-compression video coding method for video analysis using Deep Learning Tomonori Kubota, Takanori Nakao, Eiji Yoshida (Fujitsu Lab.) IMQ2019-39 IE2019-121 MVE2019-60 |
In this paper, we propose a high-compression video coding method for video analysis using Deep Learning. The method anal... [more] |
IMQ2019-39 IE2019-121 MVE2019-60 pp.121-126 |
NC, MBE |
2019-12-06 16:30 |
Aichi |
Toyohashi Tech |
Explaining Neural Networks by using a multiple tree Shunya Sasaki, Masafumi Hagiwara (Keio Univ) MBE2019-58 NC2019-49 |
The existing Neural Networks (NNs) have a problem that it is difficult to explain the reasoning process and the grounds ... [more] |
MBE2019-58 NC2019-49 pp.79-84 |
AI |
2019-11-28 13:05 |
Fukuoka |
|
A proposal of a method for analyzing causes of incorrect detection when detecting objects using Deep Learning Tomonori Kubota, Takanori Nakao, Eiji Yoshida (Fujitsu Lab.) AI2019-30 |
In this paper, we propose a method for analyzing the causes of incorrect detection / poor accuracy when detecting object... [more] |
AI2019-30 pp.1-6 |
PRMU, MI, IPSJ-CVIM [detail] |
2019-09-05 13:40 |
Okayama |
|
A method for visualizing the cause of misrecognition in object recognition using CNN Tomonori Kubota (Fujitsu Lab.), Yasuyuki Murata (FST), Yoshifumi Uehara, Akira Nakagawa (Fujitsu Lab.) PRMU2019-25 MI2019-44 |
In this paper, we propose a method for visualizing the cause of misrecognition in object recognition using CNN. By this ... [more] |
PRMU2019-25 MI2019-44 pp.99-104 |