Development of machine learning methods for materials informatics

• Project Area: Exploration of nanostructure-property relationships for materials innovation
• Project/Area Number: 16H00736
• Research Category: Grant-in-Aid for Scientific Research on Innovative Areas (Research in a proposed research area)
• Allocation Type: Single-year Grants
• Review Section: Science and Engineering
• Research Institution: The University of Tokyo
• Principal Investigator: Tsuda Koji 東京大学, 大学院新領域創成科学研究科, 教授 (90357517)
• Co-Investigator(Kenkyū-buntansha): 志賀 元紀 岐阜大学, 工学部, 准教授 (20437263) ; 鹿島 久嗣 京都大学, 情報学研究科, 教授 (80545583)
• Co-Investigator(Renkei-kenkyūsha): TAKEUCHI Ichiro 名古屋工業大学, 工学(系)研究科, 教授 (40335146)
• Project Period (FY): 2016-04-01 – 2018-03-31
• Project Status: Completed (Fiscal Year 2017)
• Budget Amount: ¥12,090,000 (Direct Cost: ¥9,300,000、Indirect Cost: ¥2,790,000); Fiscal Year 2017: ¥5,070,000 (Direct Cost: ¥3,900,000、Indirect Cost: ¥1,170,000); Fiscal Year 2016: ¥7,020,000 (Direct Cost: ¥5,400,000、Indirect Cost: ¥1,620,000)
• Keywords: マテリアルズインフォマティクス / 機械学習

Outline of Final Research Achievements

This research started from 2016 to boost the development of informatics methods in materials development. As a result of collaborations with other groups, we obtained the following achievements. (1) Statistical machine learning methods for spectrum analysis to discovering nanostructures. (2) Automatic design of materials using Monte Carlo tree search. (3) Discovery of stable compounds with recommendation algorithms.

Research Products

• [Journal Article] Informatics-Aided Raman Microscopy for Nanometric 3D Stress Characterization (2018)
  • Author(s): Wang Hongxin、Zhang Han、Da Bo、Shiga Motoki、Kitazawa Hideaki、Fujita Daisuke
  • Journal Title: The Journal of Physical Chemistry C Volume: 122 Issue: 13 Pages: 7187-7193
  • DOI: 10.1021/acs.jpcc.7b12415
  • Peer Reviewed
• [Journal Article] Exploring a potential energy surface by machine learning for characterizing atomic transport (2018)
  • Author(s): Kanamori Kenta、Toyoura Kazuaki、Honda Junya、Hattori Kazuki、Seko Atsuto、Karasuyama Masayuki、Shitara Kazuki、Shiga Motoki、Kuwabara Akihide、Takeuchi Ichiro
  • Journal Title: Physical Review B Volume: 97 Issue: 12 Pages: 125124-125124
  • DOI: 10.1103/physrevb.97.125124
  • Peer Reviewed
• [Journal Article] ChemTS: an efficient python library for de novo molecular generation (2017)
  • Author(s): Yang Xiufeng、Zhang Jinzhe、Yoshizoe Kazuki、Terayama Kei、Tsuda Koji
  • Journal Title: Science and Technology of Advanced Materials Volume: 18 Issue: 1 Pages: 972-976
  • DOI: 10.1080/14686996.2017.1401424
  • Peer Reviewed / Open Access / Int'l Joint Research
• [Journal Article] Transfer Learning to Accelerate Interface Structure Searches (2017)
  • Author(s): Oda Hiromi、Kiyohara Shin、Tsuda Koji、Mizoguchi Teruyasu
  • Journal Title: Journal of the Physical Society of Japan Volume: 86 Issue: 12 Pages: 123601-123601
  • DOI: 10.7566/jpsj.86.123601
  • NAID: 210000134600
  • Peer Reviewed / Open Access
• [Journal Article] RNA inverse folding using Monte Carlo tree search (2017)
  • Author(s): Yang Xiufeng、Yoshizoe Kazuki、Taneda Akito、Tsuda Koji
  • Journal Title: BMC Bioinformatics Volume: 18 Issue: 1 Pages: 468-468
  • DOI: 10.1186/s12859-017-1882-7
  • Peer Reviewed / Open Access
• [Journal Article] MDTS: automatic complex materials design using Monte Carlo tree search (2017)
  • Author(s): M. Dieb Thaer、Ju Shenghong、Yoshizoe Kazuki、Hou Zhufeng、Shiomi Junichiro、Tsuda Koji
  • Journal Title: Science and Technology of Advanced Materials Volume: 18 Issue: 1 Pages: 498-503
  • DOI: 10.1080/14686996.2017.1344083
  • Peer Reviewed / Open Access
• [Journal Article] Learning to Enumerate (2016)
  • Author(s): Patrick Joeger, Yukino Baba, Hisashi Kashima
  • Journal Title: Proceedings of the 25th International Conference on Artificial Neural Networks (ICANN) Volume: 1 Pages: 453-460
  • DOI: 10.1007/978-3-319-44778-0_53
  • ISBN: 9783319447773, 9783319447780
  • Peer Reviewed / Int'l Joint Research / Acknowledgement Compliant
• [Journal Article] Budgeted Stream-based Active Learning via Adaptive Submodular Maximization (2016)
  • Author(s): Kaito Fujii, Hisashi Kashima
  • Journal Title: Advances in Neural Information Processing Systems (NIPS) Volume: 1 Pages: 514-522
  • NAID: 40021033042
  • Peer Reviewed / Open Access / Acknowledgement Compliant
• [Journal Article] Matrix Factorization for Automatic Chemical Mapping from Electron Microscopic Spectral Imaging Datasets (2016)
  • Author(s): Motoki Shiga, Shunsuke Muto, Kazuyoshi Tatsumi, Koji Tsuda
  • Journal Title: Transactions of the Materials Research Society of Japan Volume: 41 Issue: 4 Pages: 333-336
  • DOI: 10.14723/tmrsj.41.333
  • NAID: 130005172417
  • ISSN: 1382-3469, 2188-1650
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Sparse Modeling of EELS and EDX Spectral Imaging Data by Nonnegative Matrix Factorization (2016)
  • Author(s): Motoki Shiga, Kazuyoshi Tatsumi, Shunsuke Muto, Koji Tsuda, Yuta Yamamoto, Toshiyuki Mori, Takayoshi Tanji
  • Journal Title: Ultramicroscopy Volume: 170 Pages: 43-59
  • DOI: 10.1016/j.ultramic.2016.08.006
  • Peer Reviewed / Int'l Joint Research / Acknowledgement Compliant
• [Journal Article] Machine-learning prediction of d-band center for metals and bimetals (2016)
  • Author(s): I. Takigawa, K. Shimizu, K. Tsuda and S. Takakusagi
  • Journal Title: RSC Advances Volume: 6 Issue: 58 Pages: 52587-52595
  • DOI: 10.1039/c6ra04345c
  • NAID: 120006219935
  • Peer Reviewed / Open Access / Acknowledgement Compliant
• [Journal Article] COMBO: An Efficient Bayesian Optimization Library for Materials Science (2016)
  • Author(s): T. Ueno, T.D. Rhone, Z. Hou, T. Mizoguchi and K. Tsuda
  • Journal Title: Materials Discovery Volume: 4 Pages: 18-21
  • DOI: 10.1016/j.md.2016.04.001
  • Peer Reviewed / Int'l Joint Research / Acknowledgement Compliant
• [Presentation] 統計的機械学習を用いたスペクトルイメージ解析 (2018)
  • Author(s): 志賀元紀
  • Organizer: 顕微ナノ・表面科学・SPM合同シンポジウム
  • Invited
• [Presentation] スペクトルデータ解析のための統計的機械学習 (2018)
  • Author(s): 志賀元紀
  • Organizer: 合同シンポジウム（「第二回 先端計測インフォマティクス・ワークショップ」「NIMS先端計測シンポジウム 2018」）
  • Invited
• [Presentation] スペクトラムイメージ解析のための統計的機械学習法 (2018)
  • Author(s): 志賀元紀
  • Organizer: 第31回日本放射光学会年会・放射光科学合同シンポジウム
  • Invited
• [Presentation] アクセプター添加BaZrO3におけるドーパントおよび酸素欠損の固溶状態に関する第一原理計算 (2018)
  • Author(s): 桑原彰秀, 志賀元紀, クレイグ フィッシャー, 森分博紀
  • Organizer: 日本金属学会2017年秋季（第161回）講演大会
• [Presentation] 統計的機械学習に基づくスペクトラムイメージ解析 (2018)
  • Author(s): 志賀元紀
  • Organizer: 2017年真空・表面科学合同講演会・：データ駆動表面科学研究部会セッション
  • Invited
• [Presentation] Automatic Spectral Imaging Analysis Based on Machine Learning (2018)
  • Author(s): Motoki Shiga, Shunsuke Muto
  • Organizer: The 8th International Workshop on Electron Energy Loss Spectroscopy and Related Techniques (EDGE 2017: Enhanced Data Generated by Electrons)
  • Int'l Joint Research
• [Presentation] 機械学習に基づくスペクトラムイメージング解析 (2017)
  • Author(s): 志賀元紀
  • Organizer: 第1回 構造イメージングと情報処理研究会
  • Place of Presentation: 仙台市 東北大学原子分子材料科学高等研究機構
  • Year and Date: 2017-02-06
• [Presentation] 統計的機械学習に基づく走査型電子顕微鏡データ解析 (2017)
  • Author(s): 志賀元紀
  • Organizer: ワークショップ「先端計測インフォマティクス 大量データ時代の情報活用」
  • Place of Presentation: つくば市 物質・材料研究機構
  • Invited
• [Presentation] Machine learning for materials discovery: catalysts, grain boundaries, superlattices and RNAs (2016)
  • Author(s): Koji Tsuda
  • Organizer: Workshop I: Machine Learning Meets Many Particle Systems
  • Place of Presentation: IPAM, UCLA, Los Angeles, USA
  • Year and Date: 2016-09-26
  • Int'l Joint Research / Invited
• [Presentation] 走査型電子顕微鏡データ解析のための非負値行列分解 (2016)
  • Author(s): 志賀元紀, 武藤俊介, 巽一厳, 津田宏治
  • Organizer: 日本応用数理学会2016年度年会
  • Place of Presentation: 北九州市北九州国際会議場
  • Year and Date: 2016-09-12
  • Invited
• [Presentation] Machine Learning for Materials Discovery: Low-LTC Compounds, GrainBoundaries and Superlattices (2016)
  • Author(s): Koji Tsuda
  • Organizer: IACS Seminar
  • Place of Presentation: Harvard University, Boston USA
  • Int'l Joint Research / Invited