Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
PRMU, MVE, VRSJ-SIG-MR, IPSJ-CVIM |
2024-01-26 10:42 |
Kanagawa |
Keio Univ. (Hiyoshi Campus) |
Investigation of lightweight methods for object recognition models of YOLOv7-tiny using a combination of various kinds of convolution modules Yuedong Li, Akira Kubota (Chuo Univ.) PRMU2023-45 |
Deep learning-based object detection techniques on images have been actively studied. Recently, lightweight and fast-edg... [more] |
PRMU2023-45 pp.32-37 |
PRMU, IBISML, IPSJ-CVIM [detail] |
2023-03-02 09:40 |
Hokkaido |
Future University Hakodate (Primary: On-site, Secondary: Online) |
Pruning for Producing Efficient DNNs: Neuron Selection and Reconstruction based on Final Layer Error Koji Kamma, Toshikazu Wada (Wakayama Univ.) PRMU2022-66 IBISML2022-73 |
Deep Neural Networks (DNNs) are dominant in the field of machine learning. However, because DNN models have large comput... [more] |
PRMU2022-66 IBISML2022-73 pp.42-47 |
PRMU |
2022-12-15 14:25 |
Toyama |
Toyama International Conference Center (Primary: On-site, Secondary: Online) |
A DNN compression method based on output error of activation functions Koji Kamma, Toshikazu Wada (Wakayama Univ.) PRMU2022-38 |
Deep Neural Networks (DNNs) are dominant in the field of machine learning. However, because DNN models have large comput... [more] |
PRMU2022-38 pp.34-39 |
PRMU, IPSJ-CVIM |
2022-05-13 10:00 |
Aichi |
Toyota Technological Institute |
Robust Small Objects Detection Using YOLO-v4 with Attention and Layer Yuedong Li (Chuo Univ) PRMU2022-1 |
Abstract: It is a difficult problem how to make traditional neural network algorithm show good adaptability to the typic... [more] |
PRMU2022-1 pp.1-5 |
IT, ISEC, RCC, WBS |
2022-03-10 11:35 |
Online |
Online |
Improving accuracy by optimizing activation function for convolutional neural network using homomorphic encryption Kohei Yagyu, Ren Takeuchi, Vo Ngoc Khoi Nguyen, Masakatsu Nishigaki, Tetsushi Ohki (Shizuoka Univ.) IT2021-92 ISEC2021-57 WBS2021-60 RCC2021-67 |
The development of secure neural network technology that performs prediction while encrypting data using homomorphic enc... [more] |
IT2021-92 ISEC2021-57 WBS2021-60 RCC2021-67 pp.58-65 |
EMM |
2022-03-08 09:55 |
Online |
(Primary: Online, Secondary: On-site) (Primary: Online, Secondary: On-site) |
[Poster Presentation]
Study on JPEG Compression Resistant Watermarking Method Trained with Quantized Activation Function Shingo Yamauchi, Masaki Kawamura (Yamaguchi Univ.) EMM2021-110 |
We propose a watermarking method that introduces a quantized activation function to acquire robustness against quantizat... [more] |
EMM2021-110 pp.95-100 |
ISEC |
2021-05-19 15:30 |
Online |
Online |
[Invited Talk]
Simple Electromagnetic Analysis Against Activation Functions of Deep Neural Networks (from AIHWS 2020) Go Takatoi, Takeshi Sugawara, Kazuo Sakiyama (UEC), Yuko Hara-Azumi (Tokyo Tech), Yang Li (UEC) ISEC2021-9 |
This invited abstract is based on the papers [1] and [2]. There are physical attacks such as side-channel attacks that a... [more] |
ISEC2021-9 p.34 |
COMP, IPSJ-AL |
2020-05-09 09:30 |
Online |
Online |
On Power and limitation of adversarial example attacks Kouichi Sakurai (Kyushu Univ.) COMP2020-5 |
A risk of adversarial example attacks which cause deep learning to make wrong decisions is getting serious even from a c... [more] |
COMP2020-5 pp.33-36 |
CCS, IN (Joint) |
2019-08-02 14:20 |
Hokkaido |
KIKI SHIRETOKO NATURAL RESORT |
Effect of shapes of activation functions on predictability in the echo state network Hanten Chang (Univ. of Tsukuba), Shinji Nakaoka (Hokkaido Univ.), Hiroyasu Ando (Univ. of Tsukuba) CCS2019-23 |
We investigate prediction accuracy for time series of Echo state networks with respect to several kinds of activation fu... [more] |
CCS2019-23 pp.27-30 |
NC |
2012-07-30 13:45 |
Shiga |
Ritsumeikan Univ. College of Information Science and Engineering |
On local analytic activation function in quaternionic hopfield neural network Teijiro Isokawa, Haruhiko Nishimura, Nobuyuki Matsui (Univ. Hyogo) NC2012-18 |
It is difficult to introduce appropriate activation functions to neural networks in the hypercomplex domains, and few fu... [more] |
NC2012-18 pp.25-30 |
OPE, EMT, LQE, PN, IEE-EMT [detail] |
2011-01-27 09:30 |
Osaka |
Osaka Univ. |
Activation Function in Optical Label Processing Using Complex-Valued Neural Network Kengo Mizote, Hiroki Kishikawa, Nobuo Goto, Shin-ichiro Yanagiya (Univ of Tokushima) PN2010-32 OPE2010-145 LQE2010-130 |
Optical neural-network circuit to recognize BPSK labels for photonic label routing is proposed, which consists of optica... [more] |
PN2010-32 OPE2010-145 LQE2010-130 pp.7-12 |
NLP |
2004-10-15 10:55 |
Niigata |
Nagaoka Univ. of Technology |
On the Activation Functions and Learning characteristics in Error-Backpropagation Learning Munenori Kuriyama, Masahiro Nakagawa (Nagaoka University of Technilogy) |
The learning speed and convergence of a neural network is deeply related to the learning algorithm, network size and act... [more] |
NLP2004-64 pp.17-22 |