河原 尊之

☆2022年12月14-16日開催(東京ビッグサイト)のSEMICON Japan「アカデミア」にて出展致します。スケーラブル全結合型イジングマシン(科研費22H01559)の成果内容です。#AIチップ

☆2022年11月8日開催(台北/台湾)のIEEE A-SSCC 2022にてM2学生が遠隔発表致します。スケーラブル全結合型イジングマシン(科研費22H01559)の成果内容です。#AIチップ
7.4 "Method of Halved Interaction Elements with Regularity Arrangement that achieves Independent Double Systems for Scalable Fully Coupled Annealing Processing"

☆2022年9月28日プレスリリース(和・英)を行いました。スケーラブル全結合型イジングマシン(科研費22H01559)の成果内容です。#AIチップ
"量子Inspired技術の新展開：スケーラブルな全結合型イジング半導体システム
～組み合わせ最適化問題求解を低消費電力かつ高速に行う技術の基礎検証に成功～"

☆2022年9月24日Microprocessors and Microsystemsにて学生筆頭論文が公開となりました。スケーラブル全結合型イジングマシン(科研費22H01559)に関するオープンアクセスの論文です。#AIチップ
"Scalable Fully Coupled Annealing Processing System and Multi-chip FPGA Implementation"
　　

With RikaDai's fully coupled Ising LSI system reported in 2020 as a basis, we developed a new scalable fully coupled Ising LSI system that combines all the basic spin cells and has excellent versatility and performance for solving the combinatorial optimization problem. Our development utilizes an epoch-making method that operates as one large-scale system by combining multiple fully coupled Ising LSI chips with an extremely small number of chip-to-chip connections and only a small amount of information transmitted between chips. A system in which 16 FPGA chips are combined is demonstrated.
本研究の一部は、日本学術振興会科学研究費補助金(課題番号 22H01559)の助成を受けて実施したものです。

　A research group led by Professor Takayuki Kawahara at Tokyo University of Science (Department of Electrical Engineering, Faculty of Engineering) has developed the world’s first artificial intelligence integrated circuit (AI chip) based on fully coupled semiconductor annealing method for solving combinatorial optimization problems. This new chip architecture implements 512 fully coupled spin states in a chip measuring 2.16 mm × 2.70 mm made using 28 nm CMOS semiconductor processing. With this chip, it is possible to instantly solve a traveling salesman problem with 22 cities, which is a task that would take 1,200 years to perform on a high-performance von Neumann CPU. Using the technique we developed, we can implement a compact high-performance system with low power requirements for office equipment and tablet terminals that can easily find optimal solutions from large numbers of combinations. This technology is expected to become very widespread in the future.
成果の一部はＮＥＤＯ委託事業の結果です：2018/10-2019/09(先導調査研究枠)"高効率・高速処理を可能とするAIチップ・次世代コンピューティングの技術開発／革新的AIエッジコンピューティング技術の開発／隣接セル間のみながら全結合を実現した汎用イジングマシンＬＳＩとＡＩエッジデータ処理システムの研究開発"

人工知能をモノに搭載しましょう。
日本が得意な"組み込みマイコン"の"ＡＩ化"でもあります。
または、どこでもＡＩ、ユビキタスＡＩ、・・・。
”AI on Things”: "Things" with advanced information-processing, artificial intelligence (AI), capability to lead to the evolution of current IoT society.

①比江嶋龍也、川島舜、柯夢南、河原尊之, "Swarm AIに基づく複数ロボットの同期行動と協調行動の有効性," 電子情報通信学会技術研究報告 (2020.1)、②Tatsuya Hiejima, Shun Kawashima, Mengnan Ke, Takayuki Kawahara, "Effectiveness of Synchronization and Cooperative Behavior of Multiple Robots Based on Swarm AI," IEEE APCCAS 2019 (2019.11)、③比江嶋龍也、河原尊之, "スウォーム人工知能に基づく掃除ロボット群の協調行動シミュレーション," 電子情報通信学会ソサイエティ大会 (2019.9)、④Akira Minamisawa, Ryoma Iimura, Takayuki Kawahara "High-speed Sparse Ising Model on FPGA," IEEE MWSCAS 2019 (2019.8)、 ⑤Yasuhiko Yoshida, Ryo Oiwa, Takayuki Kawahara "Ternary Sparse XNOR-net for FPGA Implementation," IEEE ISNE 2018 (2018.5)、⑥Yasushi Fukuda, Takayuki Kawahar "Stochastic Weights Binary Neural Networks on FPGA," IEEE ISNE 2018 (2018.5)、⑦吉田康彦、河原尊之, "IoT向け三値化ニューラルネットワークの検討," 電子情報通信学会研究会 (2017.12)、⑧Yasushi Fukuda, Zule Xu, Takayuki Kawahara "Robustness Evaluation of Restricted Boltzmann Machine against Memory and Logic Error," IEICE Transactions on Electronics (2017.12)、⑨Kenta Someya, Ryoto Ono, Takayuki Kawahara, "Novel Ising Model Using Dimension-Control for High-Speed Solver for Ising Machines," IEEE NEWCAS 2016 (2016.6)

①Best Paper Award 受賞: Kohei Koike, Kenta Suzuki, Mengnan Ke, Kenjiro Mori, Takumi Ito, Takayuki Kawahara, "Damage-Position Identification of Wooden-House Models for Structural Health Monitoring Using Machine Learning," IEEE APCCAS 2020(2020.12)、②Kenta Suzuki, Kohei Koike, Mengnan Ke, Kenjiro Mori, Takumi Ito, Takayuki Kawahara, "Improvement of Generalization Performance for Timber Health Monitoring using Machine Learning," IEEE APCCAS 2020(2020.12)、③Ryota Tanida, Atsushi Yamamoto, Noriaki Takahashi, Natsuhiko Sakiyama, Sakuya Kishi, Takayuki Kishimoto, So Hasegawa, Kenjiro Mori, Yoichiro Hashizume, Jing Ma, Takashi Nakajima, Mikio Hasegawa, Takahiro Yamamoto, Takumi Ito, Takayuki Kawahara, "Machine Learning Classification Methods Using Data of 3-Axis Acceleration Sensors Equipped with Wireless Communication Means for Locating Wooden House Structural Damage," IEEE APCCAS 2019 (2019.11)、④Noriaki Takahashi, Natsuhiko Sakiyama, Takuji Yamamoto, Sakuya Kishi, Yoichiro Hashizume, Takashi Nakajima, Takahiro Yamamoto, Mikio Hasegawa, Takumi Ito, Takayuki Kawahara, "An Evaluation of Wooden House Health Monitoring System using PVDF Piezoelectric Sensor with 3-layer Neural Network and Inverted Binary-Data Augmentation," IEEE SAMI 2019 (2019.1)

①Yuwa Kishi, Akihiro Yamada, Mengnan Ke, Takayuki Kawahara, "
Evaluation of Read Disturbance Reduction Effect by SOT-MRAM Bi-directional Read on Device Size Dependence" IEEE Intermag 2021 (2021.4)、②Yuwa Kishi, Keisuke Tabata, Mengnan Ke, Takayuki Kawahara, "Evaluation of Read Disturbance Reduction Effect by Bi-directional Read on Ferromagnetic Material Properties of SOT-MRAM," IEEE MMM 2020 (2020.11)、③Y. Kishi, M. Ke, M. Itoh, R. Kashiwa, T. Matsuzaki, T. Kawahara , "Impact of the metal/YIG interfaces in Pt/(W)/Y3Fe5O12/Gd3Ga5O12 and W/Y3Fe5O12/Gd3Ga5O12 structures for spin Seebeck effect," IEEE Intermag 2020 (2020.5)、④Atsushi Yamamoto, Takayuki Kawahara, "Evaluation of Spin Seebeck Effect in Single-crystal and Polycrystal Y3Fe5O12 (YIG)," IEEE MMM-Intermag (2019.1)、⑤Atsushi Yamamoto, Makoto Arai, Tetsuya Takimoto, Masatoshi Itoh, Takayuki Kawahara, "Evaluation of Correlation between Orientation of Y3Fe5O12 (YIG) Thin Film and Spin Seebeck Effect," 応用物理学会春季学術講演(2018.3)、⑥Atsushi Yamamoto, Makoto Arai, Tetsuya Takimoto, Masatoshi Itoh, Shizutoshi Ando, Shigeru Saito, Takayuki Kawahara, "Evaluation of Correlation Between Orientation of Y3Fe5O12(YIG) Thin Film and Spin Seebeck Effect," IEEE Intermag (2017.4)、⑧Hirokazu Kazama, Takayuki Kawahara, "Spin-Orbit Torque MRAM Read Reliability," IEEE Intermag