| Presentation | 2018-06-14 How to represent a polynomial function by Neural Networks with rectified linear units Hideaki Okazaki, Kazuya Ozawa, Kaito Isogai, Toshihiro Tachibana, Hideo Nakano, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this report we study how to represent an exact polynomial function by neural networks (NN) with rectified linear units (ReLU). We focus on a three-layer neural network that has two input nodes, four hidden processing nodes, and a single output node for a polynomial function : $alpha_0 + alpha_1 x_1 + alpha_2 x_2 + alpha_3 x_1 x_2$. We give a method to obtain the hyperparameters of the three-layer ReLU NN representing $alpha_0 + alpha_1 x_1 + alpha_2 x_2 + alpha_3 x_1 x_2$. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Neural Network / Polynomial function representation |
| Paper # | CAS2018-2,VLD2018-5,SIP2018-22,MSS2018-2 |
| Date of Issue | 2018-06-07 (CAS, VLD, SIP, MSS) |
| Conference Information | |
| Committee | CAS / SIP / MSS / VLD |
|---|---|
| Conference Date | 2018/6/14(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Hokkaido Univ. (Frontier Research in Applied Sciences Build.) |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | System and Signal Processing, etc |
| Chair | Hideaki Okazaki(Shonan Inst. of Tech.) / Shogo Muramatsu(Niigata Univ.) / Morikazu Nakamura(Univ. of Ryukyus) / Noriyuki Minegishi(Mitsubishi Electric) |
| Vice Chair | Taizo Yamawaki(Hitachi) / Naoyuki Aikawa(TUS) / Kazunori Hayashi(Osaka City Univ) / Shigemasa Takai(Osaka Univ.) / Nozomu Togawa(Waseda Univ.) |
| Secretary | Taizo Yamawaki(Shonan Inst. of Tech.) / Naoyuki Aikawa(Hitachi) / Kazunori Hayashi(Takushoku Univ.) / Shigemasa Takai(Hiroshima Univ.) / Nozomu Togawa(Toshiba) |
| Assistant | Motoi Yamaguchi(Renesas Electronics) / / Hideki Kinjo(Okinawa Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Circuits and Systems / Technical Committee on Signal Processing / Technical Committee on Mathematical Systems Science and its applications / Technical Committee on VLSI Design Technologies |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | How to represent a polynomial function by Neural Networks with rectified linear units |
| Sub Title (in English) | |
| Keyword(1) | Neural Network |
| Keyword(2) | Polynomial function representation |
| 1st Author's Name | Hideaki Okazaki |
| 1st Author's Affiliation | Shonan Institute of Technology(SIT) |
| 2nd Author's Name | Kazuya Ozawa |
| 2nd Author's Affiliation | Shonan Institute of Technology(SIT) |
| 3rd Author's Name | Kaito Isogai |
| 3rd Author's Affiliation | Shonan Institute of Technology(SIT) |
| 4th Author's Name | Toshihiro Tachibana |
| 4th Author's Affiliation | Shonan Institute of Technology(SIT) |
| 5th Author's Name | Hideo Nakano |
| 5th Author's Affiliation | Shonan Institute of Technology(SIT) |
| Date | 2018-06-14 |
| Paper # | CAS2018-2,VLD2018-5,SIP2018-22,MSS2018-2 |
| Volume (vol) | vol.118 |
| Number (no) | CAS-82,VLD-83,SIP-84,MSS-85 |
| Page | pp.pp.7-10(CAS), pp.7-10(VLD), pp.7-10(SIP), pp.7-10(MSS), |
| #Pages | 4 |
| Date of Issue | 2018-06-07 (CAS, VLD, SIP, MSS) |