Presentation | 2022-05-20 Visualization of Important Features for Classifier Decisions using Deep Image Synthesis Yushi Haku, Megumi Nakao, Tetsuya Matsuda, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | It is difficult to know the basis for the decisions of machine learning models, and it is necessary to provide a highly interpretable explanation for the model's predictions. In this study, we developed a framework for visualizing features related to the decisions of classification models based on deep image generation. Similar and adversary images are generated from the input images, and the differences between them are visualized. Conditional Variational AutoEncoder was used for image generation, aiming at image transformation considering data distribution in the feature space. The effectiveness of the proposed method was confirmed on the MNIST dataset used for handwritten character recognition, and important features in the classification of the number of fibula fragments were visualized using mandible reconstruction planning data. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Interpretability / Classifier / Deep image generarion / Mandibular reconstruction |
Paper # | SIP2022-28,BioX2022-28,IE2022-28,MI2022-28 |
Date of Issue | 2022-05-12 (SIP, BioX, IE, MI) |
Conference Information | |
Committee | SIP / BioX / IE / MI / ITE-IST / ITE-ME |
---|---|
Conference Date | 2022/5/19(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Kumamoto University Kurokami Campus |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Yukihiro Bandou(NTT) / Hitoshi Imaoka(NEC) / Kazuya Kodama(NII) / Hidekata Hontani(Nagoya Inst. of Tech.) |
Vice Chair | Toshihisa Tanaka(Tokyo Univ. Agri.&Tech.) / Takayuki Nakachi(Ryukyu Univ.) / Masatsugu Ichino(Univ. of Electro-Comm.) / Naoyuki Takada(SECOM) / Hiroyuki Bandoh(NTT) / Toshihiko Yamazaki(Univ. of Tokyo) / Hideaki Haneishi(Chiba Univ.) / Takayuki Kitasaka(Aichi Inst. of Tech.) |
Secretary | Toshihisa Tanaka(Xiaomi) / Takayuki Nakachi(Takushoku Univ.) / Masatsugu Ichino(Tokyo Univ. Agri.&Tech.) / Naoyuki Takada(KDDI Research) / Hiroyuki Bandoh(MitsubishiElectric) / Toshihiko Yamazaki(KDDI Research) / Hideaki Haneishi(Nagoya Inst. of Tech.) / Takayuki Kitasaka(Yamaguchi Univ.) / (Univ. of Hyogo) |
Assistant | Taichi Yoshida(UEC) / Seisuke Kyochi(Univ. of Kitakyushu) / Hiroyuki Suzuki(Gunma Univ) / Akihiro Hayasaka(NEC) / Shunsuke Iwamura(NHK) / Shinobu Kudo(NTT) / Hotaka Takizawa(Tsukuba Univ.) / Yoshito Otake(NAIST) |
Paper Information | |
Registration To | Technical Committee on Signal Processing / Technical Committee on Biometrics / Technical Committee on Image Engineering / Technical Committee on Medical Imaging / Technical Group on Information Sensing Technologies / Technical Group on Media Engineering |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Visualization of Important Features for Classifier Decisions using Deep Image Synthesis |
Sub Title (in English) | |
Keyword(1) | Interpretability |
Keyword(2) | Classifier |
Keyword(3) | Deep image generarion |
Keyword(4) | Mandibular reconstruction |
1st Author's Name | Yushi Haku |
1st Author's Affiliation | Kyoto University(Kyoto Univ.) |
2nd Author's Name | Megumi Nakao |
2nd Author's Affiliation | Kyoto University(Kyoto Univ.) |
3rd Author's Name | Tetsuya Matsuda |
3rd Author's Affiliation | Kyoto University(Kyoto Univ.) |
Date | 2022-05-20 |
Paper # | SIP2022-28,BioX2022-28,IE2022-28,MI2022-28 |
Volume (vol) | vol.122 |
Number (no) | SIP-28,BioX-29,IE-30,MI-31 |
Page | pp.pp.144-149(SIP), pp.144-149(BioX), pp.144-149(IE), pp.144-149(MI), |
#Pages | 6 |
Date of Issue | 2022-05-12 (SIP, BioX, IE, MI) |