| Presentation | 2021-03-03 Hybrid Sparsity in Convolutional Neural Networks Shoma Noguchi, Yukari Yamauchi, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Convolutional neural networks (CNNs) have achieved high accuracy in areas such as image classification and object detection. However, large-scale CNN models contain a lot of redundant parameters, which leads to poor generalization performance and overtraining. Alvarez et al. and Scardapane et al. applied sparse group LASSO, which is a combination of group-wise L2 regularization and element-wise L1 regularization, to various architectures. They obtained compressed networks with the same or better accuracy. Kevin et al. proposed sparse-group LASSO L0, in which the L1 regularization term of sparse-group LASSO is replaced by an L0 regularization term. In this study, we propose a hybrid sparsification method that combines sparse-group LASSO and sparse-group LASSO L0, and try to achieve higher sparsity than the conventional method while maintaining the accuracy of CNN. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Convolutional Neural Network / Sparse Modeling / Sparse Group LASSO |
| Paper # | NC2020-46 |
| Date of Issue | 2021-02-24 (NC) |
| Conference Information | |
| Committee | NC / MBE |
|---|---|
| Conference Date | 2021/3/3(3days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Neuro Computing, Medical Engineering, etc. |
| Chair | Kazuyuki Samejima(Tamagawa Univ) / Takashi Watanabe(Tohoku Univ.) |
| Vice Chair | Rieko Osu(Waseda Univ.) / Ryuhei Okuno(Setsunan Univ.) |
| Secretary | Rieko Osu(NTT) / Ryuhei Okuno(ATR) |
| Assistant | Ken Takiyama(TUAT) / Nobuhiko Wagatsuma(Toho Univ.) / Akihiro Karashima(Tohoku Inst. of Tech.) / Jun Akazawa(Meiji Univ. of Integrative Medicine) |
| 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) | Hybrid Sparsity in Convolutional Neural Networks |
| Sub Title (in English) | |
| Keyword(1) | Convolutional Neural Network |
| Keyword(2) | Sparse Modeling |
| Keyword(3) | Sparse Group LASSO |
| 1st Author's Name | Shoma Noguchi |
| 1st Author's Affiliation | Nihon University(Nihon Univ.) |
| 2nd Author's Name | Yukari Yamauchi |
| 2nd Author's Affiliation | Nihon University(Nihon Univ.) |
| Date | 2021-03-03 |
| Paper # | NC2020-46 |
| Volume (vol) | vol.120 |
| Number (no) | NC-403 |
| Page | pp.pp.21-24(NC), |
| #Pages | 4 |
| Date of Issue | 2021-02-24 (NC) |