Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
IBISML |
2022-12-23 09:20 |
Kyoto |
Kyoto University (Primary: On-site, Secondary: Online) |
Embedding stochastic differential equations into neural networks using duality Naoki Sugishita, Jun Ohkubo (Saitama Univ.) IBISML2022-51 |
Neural network training requires a large amount of data. However, sometimes we have information on the underlying equati... [more] |
IBISML2022-51 pp.54-61 |
IBISML |
2022-12-23 09:40 |
Kyoto |
Kyoto University (Primary: On-site, Secondary: Online) |
Effect of memory unit initialization on performance for function approximation Yuto Terasawa, Jun Ohkubo (Saitama Univ.) IBISML2022-52 |
Many researchers have proposed various neural network models for learning time-series data, such as RNN, LSTM, and Trans... [more] |
IBISML2022-52 pp.62-69 |
MBE, NC |
2022-12-03 16:15 |
Osaka |
Osaka Electro-Communication University |
Optimization and geometric picture of pump current in stochastic models Ryuji Sano, Jun Ohkubo (Saitama Univ.) MBE2022-41 NC2022-63 |
In stochastic models, there is a phenomenon in which an oscillating periodic external field drives the flow in one direc... [more] |
MBE2022-41 NC2022-63 pp.92-97 |
MBE, NC |
2022-12-03 16:40 |
Osaka |
Osaka Electro-Communication University |
Correlation-based discretization method of continuous variables in annealing machines Yuki Furue (Saitama Univ.), Makiko Konoshima (Fujitsu), Hirotaka Tamura (DXR Lab. Inc.), Jun Ohkubo (Saitama Univ.) MBE2022-42 NC2022-64 |
Recently, annealing hardware specialized to combinatorial optimization problems has been developed, and there are some s... [more] |
MBE2022-42 NC2022-64 pp.98-103 |
NC, MBE (Joint) |
2022-09-29 10:50 |
Miyagi |
Tohoku Univ. (Primary: On-site, Secondary: Online) |
Improvement of AdaBoost algorithm for spiking neural networks Masaya Kawaguchi, Jun Ohkubo (Saitama Univ.) NC2022-34 |
Unlike artificial neural networks (ANNs), which have been widely used recently, spiking neural networks (SNNs) have attr... [more] |
NC2022-34 pp.6-10 |
NC, MBE (Joint) |
2022-09-29 11:15 |
Miyagi |
Tohoku Univ. (Primary: On-site, Secondary: Online) |
State estimation for continuous-time stochastic systems by holonomic gradient method Riku Yamamoto, Jun Ohkubo (Saitama Univ.) NC2022-35 |
The holonomic gradient method efficiently gives us numerical values of integrals with parameters, which could be a power... [more] |
NC2022-35 pp.11-15 |
NC, MBE (Joint) |
2021-11-26 16:15 |
Online |
Online |
Deep Learning Hybrid Models for Sentiment Analysis Yunpeng Rong, Jun Ohkubo (Saitama Univ.) NC2021-30 |
Sentiment analysis (SA), which can analyze the public attitudes towards various texts, has earned increasing attention f... [more] |
NC2021-30 pp.13-17 |
MBE, NC (Joint) |
2021-10-29 11:15 |
Online |
Online |
Visualization and quantification of the difficulty of combinatorial optimization problems in Ising formulation Keiichi Soejima (Saitama Univ.), Makiko Konoshima, Hirotaka Tamura (Fujitsu), Jun Ohkubo (Saitama Univ.) NC2021-25 |
With the aim of rapidly solving combinatorial optimization problems, dedicated hardware using the Ising Model is being d... [more] |
NC2021-25 pp.40-45 |
MBE, NC (Joint) |
2021-10-29 11:40 |
Online |
Online |
A numerical study on the relationship between complexity and accuracy of neural networks based on ordinary differential equations Kaoru Esashika, Jun Ohkubo (Saitama Univ.) NC2021-26 |
In recent years, many reports have been published on deep neural networks. The residual networks have contributed to rem... [more] |
NC2021-26 pp.46-50 |
MBE, NC (Joint) |
2021-10-29 12:05 |
Online |
Online |
Examination of encoding method of Markov source in spiking neural network Kiyotaka Sekine, Jun Ohkubo (Saitama Univ.) NC2021-27 |
[more] |
NC2021-27 pp.51-56 |
MBE, NC, NLP, CAS (Joint) [detail] |
2020-10-29 16:10 |
Online |
Online |
Numerical research on effects of quantization in SNN learned by backpropagation Yumi Watanabe, Jun Ohkubo (Saitama Univ.) NC2020-14 |
There are many studies to quantize the parameters of neural networks. For example, while there are methods of quantizing... [more] |
NC2020-14 pp.29-33 |
NC, MBE (Joint) |
2020-03-06 10:45 |
Tokyo |
University of Electro Communications (Cancelled but technical report was issued) |
QUBO formulation of l1-norm for Ising-type computers Tomohiro Yokota (Saitama Univ.), Makiko Konoshima, Hirotaka Takamura (Fujitsu Labs), Jun Ohkubo (Saitama Univ./JST) NC2019-107 |
Recently, annealing hardware based on Ising-model, which includes quantum annealers, has attract many attentions. When w... [more] |
NC2019-107 pp.181-186 |
NC, MBE (Joint) |
2019-03-04 10:45 |
Tokyo |
University of Electro Communications |
Adjustment of exploratory behavior using mutual information in reinforcement learning Kaiji Koyama, Jun Ohkubo (Saitama Univ.) NC2018-51 |
One of the important problems in reinforcement learning is the
exploration-exploitation trade-off. In this research, ... [more] |
NC2018-51 pp.43-47 |
NC, MBE (Joint) |
2019-03-04 11:10 |
Tokyo |
University of Electro Communications |
Numerical experiments of QUBO formulation for ReLU-type functions Go Sato (Saitama Univ.), Makiko Koreshima, Takuya Owa, Hirotaka Tamura (Fujitsu Labs), jun Ohkubo (Saitama Univ.) NC2018-52 |
[more] |
NC2018-52 pp.49-54 |
EA, US (Joint) |
2009-01-29 16:10 |
Kyoto |
|
Acoustic imaging in indoor environments using simultaneous transmission of M-sequence signals Hiroshi Matsuo (Chiba Univ.), Junji Okubo (Tokyo Tech.), Tadashi Yamaguchi (Chiba Univ.), Hiroyuki Hachiya (Tokyo Tech.) US2008-81 EA2008-123 |
Acoustic sensing in the air has the potential to acquire information about an object such as its position, shape, materi... [more] |
US2008-81 EA2008-123 pp.49-54(US), pp.41-46(EA) |