Publications supported by the project:
2019
- Taichi Asami, Ryo Masumura, Yushi Aono, and Koichi Shinoda.
Recurrent out-of-vocabulary word detection based on distribution of features.
Computer speech & language, 58():247–259, Nov 2019.
[BibTeX▼] - Raden Mu'az Mun'im, Nakamasa Inoue, and Koichi Shinoda.
Sequence-level knowledge distillation for model compression of attention-based sequence-to-sequence speech recognition.
In ICASSP2019, 6151–6155. May 2019.
[BibTeX▼] - Kuniaki Uto, Mauro Dalla Mura, Jocelyn Chanussot, and Koichi Shinoda.
Estimation of skylight conditions based on leaf-scale wheat images.
In Images et data : méthodes d'analyse et modélisation pour l'agriculture numérique. Mar 2019.
[BibTeX▼] - Dongxiao WANG, Koichi SHINODA, and Hirokazu KAMEOKA.
A robust algorithm of phase recovery for speech enhancement.
In IEICE Technical Report, 137–142. Mar 2019.
[BibTeX▼] - Dongxiao Wang, Hirokazu Kameoka, and Koichi Shinoda.
Improving the robustness of multiple input spectrogram inversion.
In Acoustical Society of Japan (ASJ) 2019 Spring Meeting, 1307–1308. Mar 2019.
[BibTeX▼] - Tifani Warnita, Nakamasa Inoue, and Koichi Shinoda.
Detecting alzheimer's disease using gated convolutional neural network from audio data.
In IEICE Technical Report. Feb 2019.
[BibTeX▼] - H. Naganuma and R. Yokota.
Smoothing of the objective function for large scale parallel deep learning.
In The 81st National Convention of IPSJ. Fukuoka, Japan, March 14-16 2019.
[BibTeX▼] - H. Ootomo and R. Yokota.
Batched qr decomposition using tensorcores.
In The 81st National Convention of IPSJ. Fukuoka, Japan, March 14-16 2019.
[BibTeX▼] - H. Nakata, K. Osawa, and R. Yokota.
Variational inference in deep learning using natural gra- dient descent.
In The 81st National Convention of IPSJ. Fukuoka, Japan, March 14-16 2019.
[BibTeX▼] - K. Osawa, R. Yokota, C.-S. Foo, and V. Chandrasekhar.
Second order optimization for large scale parallel deep learning through analysis of the fisher information matrix.
In The 81st National Convention of IPSJ. Fukuoka, Japan, March 14-16 2019.
[BibTeX▼] - R. Yokota, K. Osawa, Y. Tsuji, Y. Ueno, and A. Naruse.
Second order optimization for large scale parallel deep learning.
In IEICE General Conference. Tokyo, Japan, March 19-22 2019.
[BibTeX▼] - H. Naganuma and R. Yokota.
Improving the generalization gap in large-batch training using noise injection.
In IEICE General Conference (poster session). Tokyo, Japan, March 19-22 2019.
[BibTeX▼] - Y. Ueno and R. Yokota.
Exhaustive study of hierarchical allreduce patterns for large messages between gpus.
In 19th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID). Larnaca, Cyprus, May 14-17 2019.
[BibTeX▼] - H. Naganuma and R. Yokota.
A performance improvement approach for second-order optimization in large mini-batch training.
In 2nd High Performance Machine Learning Workshop CCGrid2019 (HPML2019). Larnaca, Cyprus, May 14-17 2019.
[BibTeX▼] - H. Naganuma and R. Yokota.
Effectiveness of smoothing for large-batch training using natural gradient descent.
In The 3rd Cross-disciplinary Workshop on Computing Systems, Infrastruc- tures, and Programming (xSIG). Tokyo, Japan, May 27-29 2019.
[BibTeX▼] - K. Osawa, Y. Tsuji, Y. Ueno, A. Naruse, R. Yokota, and S. Matsuoka.
Second-order optimization method for large mini-batch: training resnet-50 on imagenet in 35 epochs.
In IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, USA, June 16-20 2019.
[BibTeX▼] - P. Spalthoff and R. Yokota.
Flexible and simplistic hierarchical matrix-based fast direct solver.
In The 170th Workshop on High Performance Computing (SWoPP2019). Kitami, Japan, July 24 2019.
[BibTeX▼] - H. Ootomo and R. Yokota.
Gpu implementation of tsqr using tensor cores.
In The 170th Work- shop on High Performance Computing (SWoPP2019). Kitami, Japan, July 24 2019.
[BibTeX▼] - K. Osawa, S. Swaroop, A. Jain, R. Eschenhagen, R. E. Turner, R. Yokota, and M. E. Khan.
Practical deep learning with bayesian principles.
In The 33rd Conference on Neural Information Processing Systems (NeurIPS 2019). Vancouver, Canada, Dec. 8-14 2019.
[BibTeX▼]
2018
- Nakamasa Inoue and Koichi Shinoda.
Few-shot adaptation for multimedia semantic indexing.
In Proceedings of the 26th ACM international conference on Multimedia, 1110–1118. Oct 2018.
[BibTeX▼] - Koji Okabe, Takafumi Koshinaka, and Koichi Shinoda.
Attentive statistics pooling for deep speaker embedding.
In Proc. Interspeech 2018, 2252–2256. Sep 2018.
[BibTeX▼] - Jiacen Zhang, Nakamasa Inoue, and Koich Shinoda.
I-vector transformation using conditional generative adversarial networks for short utterance speaker verification.
In Proc. Interspeech. 2018.
[BibTeX▼] - Tifani Warnita, Nakamasa Inoue, and Koichi Shinoda.
Detecting alzheimer's disease using gated convolutional neural network from audio data.
In Proc. Interspeech. 2018.
[BibTeX▼] - Thao-Minh Le, Nakamasa Inoue, and Koichi Shinoda.
A fine-to-coarse convoluational neural network for 3d human action recognition.
In Proc. British Machine Vision Conference (BMVC). 2018.
[BibTeX▼] - Thao Le Minh, Nobuyuki Shimizu, Takashi Miyazaki, and Koichi Shinoda.
Deep learning based multi-modal addressee recognition in visual scenes with utterances.
In Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI-18), 1546–1553. Jul 2018.
[BibTeX▼] - Haoyi Zhang, Conggui Liu, Nakamasa Inoue, and Koichi Shinoda.
Multi-task autoencoder for noise-robust speech recognition.
In Proc. ICASSP, 5599–5603. 2018.
[BibTeX▼] - K. A. LEE, H. Yamamoto, K. Okabe, Q. Wang, L. Guo, T. Koshinaka, J. Zhang, and K. Shinoda.
The nec-tt speaker verification system for sre'18.
In Proc. NIST 2018 Speaker Recognition Evaluation. Dec 2018.
[BibTeX▼] - Nakamasa Inoue, Chihiro Shiraishi, Aleksandr Drozd, Koichi Shinoda, Shi-wook Lee, and Alex Chichung Kot.
Vant at trecvid 2018.
In Proc. TRECVID workshop. Nov 2018.
[BibTeX▼] - Thao-Minh Le, Nakamasa Inoue, and Koichi Shinoda.
Skeleton-based human action recognition with fine-to-coarse convolutional neural network.
In Technical Reports of IEICE PRMU, 61–64. 2018.
[BibTeX▼] - Jiacen Zhang, Nakamasa Inoue, and Koich Shinoda.
Generative adversarial network based i-vector transformation for short utterance speaker verfication.
In ASJ 2018 Autumn Meeting. 2018.
[BibTeX▼] - Tifani Warnita, Nakamasa Inoue, and Koichi Shinoda.
Alzheimer's disease prediction using audio gated convolutional neural network.
In ASJ 2018 Autumn Meeting. 2018.
[BibTeX▼] - Yan Long, Nakamasa Inoue, Koichi Shinoda, Yoichi Yatsu, Ryosuke Itoh, and Nobuyuki Kawai.
Astronomical image subtraction for transient detection using cnn.
In The 21st Meeting on Image Recognition and Understanding (MIRU). Aug 2018.
[BibTeX▼] - 金井 怜, 井上 中順, 李 時旭, and 篠田 浩一.
単語分散表現を用いた動画からのイベント検出.
In 第21回 画像の認識・理解シンポジウム (MIRU). 2018.
[BibTeX▼] - R. Yokota.
Optimization methods for large scale distributed deep learning.
In IPAM Workshop I: Big Data Meets Large-Scale Computing. September 24-28 2018.
[BibTeX▼] - Hiroki Naganuma, Shun Iwase, Linsho Kaku, Hikaru Nakata, and Rio Yokota.
Hyper-parameter tuning of approximate natural gradient methods for highly parallel distributed deep learning.
In Forum on Information Technology. 2018.
[BibTeX▼] - Hiroyuki Ohtomo, Kazuki Osawa, and Rio Yokota.
Deep learning using hierarchical low-rank approximation of the fisher information matrix.
In The 80th National Convention of IPSJ. 2018.
[BibTeX▼] - Hiroyuki Otomo, Kazuki Osawa, and Rio Yokota.
Distributed learning of deep neural networks using the kronecker factorization of the fisher information matrix.
In The 163rd Workshop on High Performance Computing. 2018.
[BibTeX▼] - Yuji Kuwamura, Kazuki Osawa, and Rio Yokota.
Hyper-parameter tuning for approximate natural gradient methods.
In The 80th National Convention of IPSJ. 2018.
[BibTeX▼] - Hiroki Naganuma and Rio Yokota.
Accelerating convolutional neural networks using low precision arithmetic.
In HPC Asia, Poster Presentation. 2018.
[BibTeX▼] - Marzena Karpinska, Bofang Li, Anna Rogers, and Aleksandr Drozd.
Subcharacter information in japanese embeddings: when is it worth it?
In In Proceedings of the Workshop on Relevance of Linguistic Structure in Neural Architectures for NLP (RELNLP) 2018, to appear. ACL, 2018.
[BibTeX▼] - Bofang Li, Aleksandr Drozd, Tao Liu, and Xiaoyong Du.
Subword-level composition functions for learning word embeddings.
In Proceedings of The 2nd Workshop on Subword and Character level models in NLP (SCLeM), 38–48. ACL, 2018.
[full text] [BibTeX▼] - Jian Guo, Akihiro Nomura, Ryan Barton, Haoyu Zhang, and Satoshi Matsuoka.
Machine learning predictions for underestimation of job runtime on hpc system.
In Asian Conference on Supercomputing Frontiers, 179–198. Springer, 2018.
[BibTeX▼] - Y. Oyama, T. Ben-Nun, T. Hoefler, and S. Matsuoka.
µ-cuDNN: Accelerating Deep Learning Frameworks with Micro-Batching.
ArXiv e-prints, April 2018.
arXiv:1804.04806.
[BibTeX▼] - Arie Wahyu Wijayanto and Tsuyoshi Murata.
Pre-emptive spectral graph protection strategies on multiplex social networks.
Applied Network Science, 3(1):5, Apr 2018.
[BibTeX▼] - Arie Wahyu Wijayanto and Tsuyoshi Murata.
Learning adaptive graph protection strategy on dynamic networks via reinforcement learning.
In Proceedings of The 2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI 2018), WI 2018, to appear. Dec 2018.
[BibTeX▼] - Jun Jin Choong, Xin Liu, and Tsuyoshi Murata.
Learning community structure with variational autoencoder.
In Proceedings of The 2018 IEEE International Conference on Data Mining (ICDM 2018). Nov 2018.
[BibTeX▼] - Kaushalya Madhawa and Tsuyoshi Murata.
Exploring partially observed networks with nonparametric bandits.
In Proceedings of The 7th International Conference on Complex Networks and their Applications (CNA 2018). Dec 2018.
[BibTeX▼] - Arie Wahyu Wijayanto, Jun Jin Choong, Kaushalaya Madhawa, and Tsuyoshi Murata.
Robustness of compressed convolutional neural networks.
In Proceedings of The 2018 IEEE Workshop on Big Data for CyberSecurity (BigData 2018). Dec 2018.
[BibTeX▼]
2017
- Conggui Liu, Nakamasa Inoue, and Koichi Shinoda.
A unified network for multi-speaker speech recognition with multi-channel recordings.
In Proc. APSIPA. 2017.
[BibTeX▼] - Yuki Yasui, Nakamasa Inoue, Koji Iwano, and Koichi Shinoda.
Multimodal speech recognition using mouth images from depth camera.
In Proc. APSIPA. 2017.
[BibTeX▼] - Mengxi Lin, Nakamasa Inoue, and Koichi Shinoda.
Ctc network with statistical language modeling for action sequence recognition in videos.
In Proc. ACM Multimedia Thematic Workshop. 2017.
[BibTeX▼] - Nakamasa Inoue, Ryosuke Yamamoto, Na Rong, Satoshi Kanai, Junsuke Masada, Chihiro Shiraishi, Shi-wook Lee, and Koichi Shinoda.
Tokyotech-aist at trecvid 2017 multimedia event detection using deep cnns and zero-shot classifiers.
In Proc. TRECVID workshop. 2017.
[BibTeX▼] - Mengxi Lin, Nakamasa Inoue, and Koichi Shinoda.
Action sequence recognition in videos by combining a ctc network with a statistical language model.
In Technical Reports of IEICE PRMU. 2017.
[BibTeX▼] - 安井 勇樹, 岩野 公司, 井上 中順, and 篠田 浩一.
口唇深度画像を利用したディープオートエンコーダに基づくマルチモーダル音声認識.
In 日本音響学会2017年秋季研究発表会講演論文集. 2017.
[BibTeX▼] - Conggui Liu, Nakamasa Inoue, and Koichi Shinoda.
Joint training of speaker separation and speech recognit ion based on deep learning.
In ASJ 2017 Autumn Meeting. 2017.
[BibTeX▼] - 安井 勇樹, 岩野 公司, 井上 中順, and 篠田 浩一.
口唇の深度画像を用いたディープオートエンコーダによるマルチモーダル音声認識.
In 情報処理学会研究報告 SLP. 2017.
[BibTeX▼] - 西 史人, 井上 中順, 岩野 公司, and 篠田 浩一.
話者認識と顔画像認識を用いた映像におけるマルチモーダル人物同定.
In 日本音響学会2017年春季研究発表会講演論文集. 2017.
[BibTeX▼] - Conggui Liu, Nakamasa Inoue, and Koichi Shinoda.
Speaker separation in multi-channel environment using deep learning.
In Technical Reports of IPSJ SLP. 2017.
[BibTeX▼] - Hiroki Naganuma and Rio Yokota.
Verification of low-precision arithmetic for the acceleration of convolutional neural networks.
In GTC Japan, Poster Presentation. 2017.
[BibTeX▼] - Akira Sekiya, Kazuki Oosawa, Hiroki Naganuma, and Rio Yokota.
Acceleration of matrix multiplication in deep learning using low-rank approximation.
In 158th Research Presentation Seminar in High Performance Computing. 2017.
[BibTeX▼] - Kazuki Oosawa, Akira Sekiya, Hiroki Naganuma, and Rio Yokota.
Acceleration of convoluational neural networks using low-rank factorization.
In Workshop on Pattern Recognition and Media Understanding. 2017.
[BibTeX▼] - Kazuki Osawa, Akira Sekiya, Hiroki Naganuma, and Rio Yokota.
Accelerating convolutional neural networks using low-rank approximation.
In Proceedings of the 22nd Conference of Japan Computational Engineering Society. 2017.
[BibTeX▼] - Kazuki Osawa, Akira Sekiya, Hiroki Naganuma, and Rio Yokota.
Accelerating matrix multiplication in deep learning by using low-rank approximation.
In The 2017 International Conference on High Performance Computing & Simulation. Genoa, Italy, 2017.
[BibTeX▼] - Kazuki Osawa and Rio Yokota.
Evaluating the compression efficiency of the filters in convolutional neural networks.
In Artificial Neural Networks and Machine Learning - ICANN 2017, volume 10614 of Lecture Notes in Computer Science, 459–466. Springer, 2017.
[BibTeX▼] -
大山 洋介, 野村 哲弘, 佐藤 育郎, and 松岡 聡.
ディープラーニングのデータ並列学習における少精度浮動小数点数を用いた通信量の削減.
IPSJ SIG Technical Report, Vol. 2017-HPC-158, pages 1–10, 2017.
[BibTeX▼] - Yosuke Oyama, Tal Ben-Nun, Torsten Hoefler, and Satoshi Matsuoka.
Less is More: Accelerating Deep Neural Networks with Micro-Batching.
IPSJ SIG Technical Report, Vol. 2017-HPC-162, pages 1–9, 2017.
[BibTeX▼] - Arie Wahyu Wijayanto and Tsuyoshi Murata.
Flow-aware vertex protection strategy on large social networks.
In Proceedings of The 2017 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2017), ASONAM 2017, 58–63. Aug 2017.
[BibTeX▼]
2016
- Yosuke Oyama, Akihiro Nomura, Ikuro Sato, Hiroki Nishimura, Yukimasa Tamatsu, and Satoshi Matsuoka.
Predicting Statistics of Asynchronous SGD Parameters for a Large-Scale Distributed Deep Learning System on GPU Supercomputers.
In 2016 IEEE International Conference on Big Data (Big Data), 66–75. 2016.
[BibTeX▼]