Presentation | 2019-03-04 Towards understandable deep learning in stacked autoencoders Masumi Ishikawa, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Recent progress of deep learning(DP) is remarkable and its recognition ability is said to surpass that of humans. The acquired features, however, are not clear to humans. In many cases, basis of judgement by DP is not known. In 1990s the author proposed l1-norm to make neural networks understandable. In DP, nonlinear in nature, l1-norm alone is not enough to understand the resulting model due to distributed representations. The present paper proposes to extend the previous proposal by jointly using selective l1-norm, hidden neurons activation regularization and so forth, and applies it to real data to verify that sparsity obtained and meaning of hidden neurons attached are effective. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Deep learning / Stacked autoencoders / Understandable AI / l1- norm / Neural networks |
Paper # | NC2018-62 |
Date of Issue | 2019-02-25 (NC) |
Conference Information | |
Committee | NC / MBE |
---|---|
Conference Date | 2019/3/4(3days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | University of Electro Communications |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Yutaka Hirata(Chubu Univ.) / Masaki Kyoso(TCU) |
Vice Chair | Hayaru Shouno(UEC) / Taishin Nomura(Osaka Univ.) |
Secretary | Hayaru Shouno(Nagoya Univ.) / Taishin Nomura(NAIST) |
Assistant | Keiichiro Inagaki(Chubu Univ.) / Takashi Shinozaki(NICT) / Takumi Kobayashi(YNU) / Yasuyuki Suzuki(Osaka Univ.) |
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) | Towards understandable deep learning in stacked autoencoders |
Sub Title (in English) | |
Keyword(1) | Deep learning |
Keyword(2) | Stacked autoencoders |
Keyword(3) | Understandable AI |
Keyword(4) | l1- norm |
Keyword(5) | Neural networks |
1st Author's Name | Masumi Ishikawa |
1st Author's Affiliation | Kyushu Institute of Technology(Kyutech) |
Date | 2019-03-04 |
Paper # | NC2018-62 |
Volume (vol) | vol.118 |
Number (no) | NC-470 |
Page | pp.pp.99-104(NC), |
#Pages | 6 |
Date of Issue | 2019-02-25 (NC) |