| Presentation | 2021-06-24 A Study of Ensemble Learning for Randomly Weighted Neural Network Yasuyuki Okoshi, Kazutoshi Hirose, Kota Ando, Kazushi Kawamura, Thiem Van Chu, Masato Motomura, Jaehoon Yu, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Recent research on deep learning shows the possibility of building neural networks by learning connection existences instead of weights, in which weights are random numbers and do not change. Connection-based learning can significantly reduce memory resources to store the huge number of weights. In this research, we adopt various ensemble learning methods to the randomly weighted neural networks for improving the trade-off between computational cost and accuracy. On CIFAR-100, our ensembled ResNet18 model achieved 3.0% higher accuracy than the original randomly weighted ResNet18 model. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | machine learning / deep learning / ensemble learning / neural network / image classification |
| Paper # | SIS2021-7 |
| Date of Issue | 2021-06-17 (SIS) |
| Conference Information | |
| Committee | SIS / IPSJ-AVM |
|---|---|
| Conference Date | 2021/6/24(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Intelligent Multimedia Systems, Applied Embedded Systems, Three-Dimensional Image Technology (3DIT), etc. |
| Chair | Noriaki Suetake(Yamaguchi Univ.) / Hiroyuki Kasai(Waseda Univ.) |
| Vice Chair | Tomoaki Kimura(Kanagawa Inst. of Tech.) / Naoto Sasaoka(Tottori Univ.) |
| Secretary | Tomoaki Kimura(National Inst. of Tech., Ube College) / Naoto Sasaoka(NTT) / (NTT) |
| Assistant | Soh Yoshida(Kansai Univ.) / Yoshiaki Makabe(Kanagawa Inst. of Tech.) |
| Paper Information | |
| Registration To | Technical Committee on Smart Info-Media Systems / Special Interest Group on Audio Visual and Multimedia Information Processing |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A Study of Ensemble Learning for Randomly Weighted Neural Network |
| Sub Title (in English) | |
| Keyword(1) | machine learning |
| Keyword(2) | deep learning |
| Keyword(3) | ensemble learning |
| Keyword(4) | neural network |
| Keyword(5) | image classification |
| 1st Author's Name | Yasuyuki Okoshi |
| 1st Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
| 2nd Author's Name | Kazutoshi Hirose |
| 2nd Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
| 3rd Author's Name | Kota Ando |
| 3rd Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
| 4th Author's Name | Kazushi Kawamura |
| 4th Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
| 5th Author's Name | Thiem Van Chu |
| 5th Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
| 6th Author's Name | Masato Motomura |
| 6th Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
| 7th Author's Name | Jaehoon Yu |
| 7th Author's Affiliation | Tokyo Institute of Technology(Tokyo Tech) |
| Date | 2021-06-24 |
| Paper # | SIS2021-7 |
| Volume (vol) | vol.121 |
| Number (no) | SIS-73 |
| Page | pp.pp.37-42(SIS), |
| #Pages | 6 |
| Date of Issue | 2021-06-17 (SIS) |