Presentation 2019-12-06
Explaining Neural Networks by using a multiple tree
Shunya Sasaki, Masafumi Hagiwara,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The existing Neural Networks (NNs) have a problem that it is difficult to explain the reasoning process and the grounds for the output. We propose multi-tree NNs. The proposed network clarifies how input elements influence each other and lead to output. The explanation is made by referring to the intermediate output distribution of the learning data. In the case of correct output, the similarity between each intermediate node output and the distribution during learning. In the case of error output, the tree structure is searched from the output layer to the input layer. In this way, a highly explanatory NNs are realized.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) explainable AI / white box / tree
Paper # MBE2019-58,NC2019-49
Date of Issue 2019-11-29 (MBE, NC)

Conference Information
Committee NC / MBE
Conference Date 2019/12/6(1days)
Place (in Japanese) (See Japanese page)
Place (in English) Toyohashi Tech
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Hayaru Shouno(UEC) / Taishin Nomura(Osaka Univ.)
Vice Chair Kazuyuki Samejima(Tamagawa Univ) / Takashi Watanabe(Tohoku Univ.)
Secretary Kazuyuki Samejima(NAIST) / Takashi Watanabe(NTT)
Assistant Takashi Shinozaki(NICT) / Ken Takiyama(TUAT) / Yasuyuki Suzuki(Osaka Univ.) / Akihiro Karashima(Tohoku Inst. of Tech.)

Paper Information
Registration To Technical Committee on Neurocomputing / Technical Committee on ME and Bio Cybernetics
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Explaining Neural Networks by using a multiple tree
Sub Title (in English)
Keyword(1) explainable AI
Keyword(2) white box
Keyword(3) tree
1st Author's Name Shunya Sasaki
1st Author's Affiliation Keio University(Keio Univ)
2nd Author's Name Masafumi Hagiwara
2nd Author's Affiliation Keio University(Keio Univ)
Date 2019-12-06
Paper # MBE2019-58,NC2019-49
Volume (vol) vol.119
Number (no) MBE-327,NC-328
Page pp.pp.79-84(MBE), pp.79-84(NC),
#Pages 6
Date of Issue 2019-11-29 (MBE, NC)