Journal Paper

• Yosuke Oyama, Naoya Maruyama, Nikoli Dryden, Erin McCarthy, Peter Harrington, Jan Balewski, Satoshi Matsuoka, Peter Nugent, Brian Van Essen. The Case for Strong Scaling in Deep Learning: Training Large 3D CNNs with Hybrid Parallelism, IEEE Transactions on Parallel & Distributed Systems (TPDS), vol. 32, no. 7, pp. 1641-1652, July 2021.

International Conference (Reviewed)

• Jens Domke, Emil Vatai, Alexsandr Drozd, Peng Chen, Yosuke Oyama, Lingqi Zhang, Shweta Salaria, Daichi Mukunoki, Artur Podobas, Mohamed Wahib, Satoshi Matsuoka. Matrix Engines for High Performance Computing: A Paragon of Performance or Grasping at Straws?, International Parallel and Distributed Processing Symposium (IPDPS 2021), May 2021.
• Yosuke Oyama, Naoya Maruyama, Nikoli Dryden, Peter Harrington, Jan Balewski, Satoshi Matsuoka, Marc Snir, Peter Nugent, Brian Van Essen. Toward Training a Large 3D Cosmological CNN with Hybrid Parallelization, 48th International Conference on Parallel Processing (ICPP 2019), Aug. 2019.
• Yosuke Oyama, Tal Ben-Nun, Torsten Hoefler, Satoshi Matsuoka. u-cuDNN: Accelerating Deep Learning Frameworks with Micro-Batches, GPU Technology Conference 2019 (GTC2019), Mar. 2019.
• Yosuke Oyama, Tal Ben-Nun, Torsten Hoefler, Satoshi Matsuoka. Accelerating Deep Learning Frameworks with Micro-batches, IEEE Cluster 2018, Sept. 2018.
• Ikuro Sato, Ryo Fujisaki, Yosuke Oyama, Akihiro Nomura, Satoshi Matsuoka. Asynchronous, data-parallel deep convolutional neural network training with linear prediction model for parameter transition, The 24th International Conference On Neural Information Processing (ICONIP 2017), International Conference on Neural Information Processing, volume 10635, pp. 305-314, Nov. 2017.
• Yosuke Oyama, Akihiro Nomura, Ikuro Sato, Hiroki Nishimura, Yukimasa Tamatsu, Satoshi Matsuoka. Predicting Probabilistic Parameters of a Large-Scale Asynchronous SGD Deep Learning System, GPU Technology Conference 2017 (GTC 2017), May 2017.
• Yosuke Oyama, Akihiro Nomura, Ikuro Sato, Hiroki Nishimura, Yukimasa Tamatsu, Satoshi Matsuoka. Predicting Statistics of Asynchronous SGD Parameters for a Large-Scale Distributed Deep Learning System on GPU Supercomputers, 2016 IEEE International Conference on Big Data (IEEE BigData 2016), Dec. 2016.
• Yosuke Oyama, Akihiro Nomura, Ikuro Sato, Hiroki Nishimura, Yukimasa Tamatsu, SATOSHI MATSUOKA. Training Condition Conscious Performance Modeling of an Asynchronous Data-Parallel Deep Learning System, ACM Symposium on High-Performance Parallel and Distributed Computing, May 2016.

International Conference (Not reviewed / Unknown)

• Yosuke Oyama, Naoya Maruyama, Nikoli Dryden, Peter Harrington, Jan Balewski, Satoshi Matsuoka, Marc Snir, Peter Nugent, Brian Van Essen. Toward Training a Large 3D Cosmological CNN with Hybrid Parallelization, The 1st Workshop on Parallel and Distributed Machine Learning 2019 (PDML'19), Aug. 2019.
• Yosuke Oyama, Hiroki Ohtsuji, Jun Kato, Kosuke Suzuki, Mitsuru Sato, Eiji Yoshida. Partially-Decompressible Dictionary Based Compression Format for All Flash Array, 1st Joint International Workshop on Parallel Data Storage & Data Intensive Scalable Computing Systems (PDSW-DISCS’16), Nov. 2016.

Domestic Conference (Not reviewed / Unknown)

• Yosuke Oyama, Naoya Maruyama, Nikoli Dryden, Peter Harrington, Jan Balewski, Satoshi Matsuoka, Marc Snir, Peter Nugent, Brian Van Essen. Toward Training a Large 3D Cosmological CNN with Hybrid Parallelization, 第170回ハイパフォーマンスコンピューティング研究発表会, July 2019.
• Toshiki Tsuchikawa, Toshio Endo, Akihiro Nomura, Masaaki Kondo, Yosuke Oyama, SATOSHI MATSUOKA. メモリアクセスデータを用いた機械学習によるアプリケーションの類型化, 並列/分散/協調処理に関するサマーワークショップ(SWoPP2019)， 情報処理学会研究報告, 2019-HPC-170 No.12, July 2019.
• Toshiki Tsuchikawa, Yosuke Oyama, Akihiro Nomura, SATOSHI MATSUOKA. 機械学習による計算機トレースの自動生成, 並列/分散/協調処理に関するサマーワークショップ (SWoPP2018), Aug. 2018.
• Keita Yashima, Yosuke Oyama, SATOSHI MATSUOKA. 深層学習におけるBatchNormalization使用時の計算時間と精度の関係性, 並列/分散/協調処理に関するサマーワークショップ (SWoPP2018), July 2018.
• Yosuke Oyama, Tal Ben-Nun, Torsten Hoefler, Satoshi Matsuoka. Less is More: Accelerating Deep Neural Networks with Micro-Batching, 第162回ハイパフォーマンスコンピューティング研究発表会, Dec. 2017.
• Yosuke Oyama, Akihiro Nomura, 佐藤育郎, SATOSHI MATSUOKA. 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, 第158回ハイパフォーマンスコンピューティング研究発表会, Mar. 2017.
• Yosuke Oyama, Akihiro Nomura, 佐藤育郎, 西村裕紀, 玉津幸政, SATOSHI MATSUOKA. 学習条件を考慮した大規模非同期ディープラーニングシステムの性能モデリング, 並列/分散/協調処理に関するサマーワークショップ(SWoPP2016), Aug. 2016.

Other Publication

• Jens Domke, Emil Vatai, Alexsandr Drozd, Peng Chen, Yosuke Oyama, Lingqi Zhang, Shweta Salaria, Daichi Mukunoki, Artur Podobas, Mohamed Wahib, Satoshi Matsuoka. Matrix Engines for High Performance Computing: A Paragon of Performance or Grasping at Straws?, Oct. 2020.  Official location  
• Yosuke Oyama, Naoya Maruyama, Nikoli Dryden, Erin McCarthy, Peter Harrington, Jan Balewski, Satoshi Matsuoka, Peter Nugent, Brian Van Essen. The Case for Strong Scaling in Deep Learning: Training Large 3D CNNs with Hybrid Parallelism, July 2020.  Official location  
• Yosuke Oyama, Tal Ben-Nun, Torsten Hoefler, Satoshi Matsuoka. μ-cuDNN, July 2018.  Official location  
• Yosuke Oyama, Tal Ben-Nun, Torsten Hoefler, SATOSHI MATSUOKA. μ-cuDNN: Accelerating Deep Learning Frameworks with Micro-Batching, Apr. 2018.  Official location  

Patent

• SATOSHI MATSUOKA, Akihiro Nomura, Yosuke Oyama. 学習システムおよび学習方法. Patent. Registered. 国立大学法人東京工業大学, 株式会社デンソーアイティーラボラトリ. 2016/12/27. 特願2016-253169. 2018/07/05. 特開2018-106489. 特許第6704583号. 2020/05/15 2020.
• SATOSHI MATSUOKA, Akihiro Nomura, Yosuke Oyama. 予測装置、予測方法および予測プログラム. Patent. Registered. 国立大学法人東京工業大学, 株式会社デンソー, 株式会社デンソーアイティーラボラトリ. 2016/07/29. 特願2016-150221. 2018/02/01. 特開2018-018422. 特許第6635265号. 2019/12/27 2019.

Degree

• Hierarchical Hybrid Parallel Training of Large-Scale Convolutional Neural Networks,  Exam Summary,  Doctor (Science),  Tokyo Institute of Technology,  2021/03/26,   
• Hierarchical Hybrid Parallel Training of Large-Scale Convolutional Neural Networks,  Summary,  Doctor (Science),  Tokyo Institute of Technology,  2021/03/26,   
• Hierarchical Hybrid Parallel Training of Large-Scale Convolutional Neural Networks,  Thesis,  Doctor (Science),  Tokyo Institute of Technology,  2021/03/26,   

[ Save as BibTeX ]    [ Paper, Presentations, Books, Others, Degrees: Save as CSV ]    [ Patents: Save as CSV ]