| Presentation | 2018-12-13 Extracting rules from convolutional neural networks Hiroshi Tsukimoto, Yuya Sato, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | To understand the inner structures of convolutional neural networks, several techniques of visualization have been developed. However, few techniques have been developed to extract features from convolutional neural networks. This paper presents a method of extracting features from convolutional neural networks by extracting rules from fully connected layers. The rule extraction method developed by one of the authors was adopted. The inputs of the rules are the outputs of convolutional layers. To understand the rules, the images maximizing the outputs of convolutional layers are needed, which are calculated by Smoothgrad. The method was applied to MNIST data. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | convolutional neural networks / feature extraction / rule extraction / Approximation method / SmoothGrad / MNIST |
| Paper # | PRMU2018-79 |
| Date of Issue | 2018-12-06 (PRMU) |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2018/12/13(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Shinichi Sato(NII) |
| Vice Chair | Yoshihisa Ijiri(Omron) / Toru Tamaki(Hiroshima Univ.) |
| Secretary | Yoshihisa Ijiri(NEC) / Toru Tamaki(Osaka Univ.) |
| Assistant | Go Irie(NTT) / Yoshitaka Ushiku(OSX) |
| Paper Information | |
| Registration To | Technical Committee on Pattern Recognition and Media Understanding |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Extracting rules from convolutional neural networks |
| Sub Title (in English) | |
| Keyword(1) | convolutional neural networks |
| Keyword(2) | feature extraction |
| Keyword(3) | rule extraction |
| Keyword(4) | Approximation method |
| Keyword(5) | SmoothGrad |
| Keyword(6) | MNIST |
| 1st Author's Name | Hiroshi Tsukimoto |
| 1st Author's Affiliation | Tokyo Denki University(Tokyo Denki Univ.) |
| 2nd Author's Name | Yuya Sato |
| 2nd Author's Affiliation | Tokyo Denki University(Tokyo Denki Univ.) |
| Date | 2018-12-13 |
| Paper # | PRMU2018-79 |
| Volume (vol) | vol.118 |
| Number (no) | PRMU-362 |
| Page | pp.pp.23-28(PRMU), |
| #Pages | 6 |
| Date of Issue | 2018-12-06 (PRMU) |