Updated on 2025/02/27

Information

 

写真a

 
YOSHIZOE KAZUKI
 
Organization
Research Institute for Information Technology Section of Advanced Computational Science Professor
Research Institute for Information Technology (Concurrent)
School of Sciences Department of Physics(Concurrent)
Graduate School of Information Science and Electrical Engineering Department of Information Science and Technology(Concurrent)
Joint Graduate School of Mathematics for Innovation (Concurrent)
Title
Professor
Contact information
メールアドレス
Tel
0928022653
Profile
中心となる研究テーマとしてグラフ探索アルゴリズムや探索などの複雑なアルゴリズムの大規模並列化に取り組んでいる。 コンピュータ囲碁に関連するアルゴリズムやより広くゲームAIもテーマとしている。 さらに探索と機械学習の応用として化合物など(いわゆるマテリアルインフォマティクス)や遺伝子解析(バイオインフォマティクス)などもテーマとしている。 教育としてはシステム情報科学府などを兼務し、グラフ探索などに関係する科目を担当予定。 情報基盤研究開発センターでスーパーコンピュータ関連の業務にも取り組む。

Research Areas

  • Informatics / High performance computing

  • Informatics / Intelligent informatics

Degree

  • Ph.D. (Computer Science)

Research History

  • Kyushu University Research Institute for Information Technology Professor 

    2021.10 - Present

      More details

  • 2003年〜2005年 株式会社 富士通研究所 無線通信に関する研究開発。 2008年〜2010年 JST ERATO-SORST量子情報システムアーキテクチャ 研究員 量子コンピュータに関する研究に取り組む。 2012年〜2015年 JST ERATO湊離散構造処理系プロジェクト 研究員 探索アルゴリズムの並列化及び機械学習応用に取り組む。 2017年〜2021年 理化学研究所 革新知能統合研究センター ユニットリーダー 「探索と並列計算ユニット」「計算支援運用ユニット」のリーダーを兼務。 機械学習+探索+並列計算アルゴリズムの理論と応用に取り組む。 同センターのAI用途スーパーコンピュータRAIDENシステムの構築・運用も担当。   

    2003年〜2005年 株式会社 富士通研究所 無線通信に関する研究開発。 2008年〜2010年 JST ERATO-SORST量子情報システムアーキテクチャ 研究員 量子コンピュータに関する研究に取り組む。 2012年〜2015年 JST ERATO湊離散構造処理系プロジェクト 研究員 探索アルゴリズムの並列化及び機械学習応用に取り組む。 2017年〜2021年 理化学研究所 革新知能統合研究センター ユニットリーダー 「探索と並列計算ユニット」「計算支援運用ユニット」のリーダーを兼務。 機械学習+探索+並列計算アルゴリズムの理論と応用に取り組む。 同センターのAI用途スーパーコンピュータRAIDENシステムの構築・運用も担当。

  • 2006年〜2008年 中央大学 研究開発機構 専任研究員(機構助教) 情報セキュリティの研究に従事。特にバイオメトリック認証の研究など。 2010年〜2012年 東京大学大学院 情報理工学系研究科 助教 並列計算の研究、特に探索アルゴリズムの並列化に取り組む。アセンブリプログラミング、Linuxプログラミングの講義を担当。 2015年〜2016年 東京大学大学院 新領域創成科学研究科 特任研究員 バイオインフォマティクス、材料科学への並列探索アルゴリズムの応用 2017年〜2021年 東京農工大学 客員准教授 2021年〜 九州大学 情報基盤研究開発センター 教授   

Research Interests・Research Keywords

  • Research theme: 量子計算

    Keyword: 量子計算

    Research period: 2024

  • Research theme: 探索

    Keyword: 探索

    Research period: 2024

  • Research theme: 情報セキュリティ

    Keyword: 情報セキュリティ

    Research period: 2024

  • Research theme: 囲碁

    Keyword: 囲碁

    Research period: 2024

  • Research theme: 人工知能

    Keyword: 人工知能

    Research period: 2024

  • Research theme: バイオメトリクス

    Keyword: バイオメトリクス

    Research period: 2024

  • Research theme: ゲーム情報学

    Keyword: ゲーム情報学

    Research period: 2024

  • Research theme: Game Tree Search

    Keyword: Game Tree Search

    Research period: 2024

  • Research theme: Biometrics

    Keyword: Biometrics

    Research period: 2024

  • Research theme: Artificial Intelligence

    Keyword: Artificial Intelligence

    Research period: 2024

  • Research theme: Solving real-world problems including, but not limited to, chemistry and material science using machine learning and graph search.

    Keyword: Graph Search, Machine Learning, Chemical Compounds, Material Science

    Research period: 2015.4

  • Research theme: Large-Scale Parallelization of Graph Search Algorithms

    Keyword: Graph Search Algorithms, Distributed Memory Parallelization

    Research period: 2010.5

  • Research theme: Game-playing algorithms using graph search and machine learning

    Keyword: Game-AI, Search Algorithms, Machine Learning

    Research period: 2005.4

Awards

  • Best Paper Award, Advances in Computer Games 13 (ACG13)

    2011.11   International Computer Games Association (ICGA)   Accelerated UCT and Its Application to Two-Player Games

Papers

  • Practical Massively Parallel Monte-Carlo Tree Search Applied to Molecular Design. Reviewed

    Xiufeng Yang, Tanuj Kr Aasawat, Kazuki Yoshizoe

    The Ninth International Conference on Learning Representations (ICLR2021)   2021.5

     More details

    Language:English   Publishing type:Research paper (other academic)  

  • Scalable Distributed Monte Carlo Tree Search Reviewed

    Kazuki Yoshizoe, Akihiro Kishimoto, Tomoyuki Kaneko, Haruhiro Yoshimoto, Yutaka Ishikawa

    Proceedings of The Fourth Annual Symposium on Combinatorial Search (SoCS2011)   4   180 - 187   2011.7

     More details

    Language:English   Publishing type:Research paper (other academic)  

  • Lambda Depth-first Proof Number Search and its Application to Go Reviewed

    Kazuki Yoshizoe, Akihiro Kishimoto, Martin Mueller

    20TH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI-2007)   2404 - 2409   2007.1

     More details

    Language:English   Publishing type:Research paper (other academic)  

    Thomsen's lambda search and Nagai's depth-first proof-number (DFPN) search are two powerful but very different AND/OR tree search algorithms. Lambda Depth-First Proof Number search (LDFPN) is a novel algorithm that combines ideas from both algorithms. lambda search can dramatically reduce a search space by finding different levels of threat sequences. DFPN employs the notion of proof and disproof numbers to expand nodes expected to be easiest to prove or disprove. The method was shown to be effective for many games. Integrating lambda order with proof and disproof numbers enables LDFPN to select moves more effectively, while preserving the efficiency of DFPN. LDFPN has been implemented for capturing problems in Go and is shown to be more efficient than DFPN and more robust than an algorithm based on classical lambda search.

  • ChemTSv2: Functional molecular design using de novo molecule generator

    Shoichi Ishida, Tanuj Aasawat, Masato Sumita, Michio Katouda, Tatsuya Yoshizawa, Kazuki Yoshizoe, Koji Tsuda, Kei Terayama

    WILEY INTERDISCIPLINARY REVIEWS-COMPUTATIONAL MOLECULAR SCIENCE   13 ( 6 )   2023.7   ISSN:1759-0876 eISSN:1759-0884

     More details

    Language:English   Publishing type:Research paper (scientific journal)   Publisher:WILEY  

    Designing functional molecules is the prerogative of experts who have advanced knowledge and experience in their fields. To democratize automatic molecular design for both experts and nonexperts, we introduce a generic open-sourced framework, ChemTSv2, to design molecules based on a de novo molecule generator equipped with an easy-to-use interface. Besides, ChemTSv2 can easily be integrated with various simulation packages, such as Gaussian 16 package, and supports a massively parallel exploration that accelerates molecular designs. We exhibit the potential of molecular design with ChemTSv2, including previous work, such as chromophores, fluorophores, drugs, and so forth. ChemTSv2 contributes to democratizing inverse molecule design in various disciplines relevant to chemistry.This article is categorized under:Data Science > Databases and Expert SystemsData Science > Artificial Intelligence/Machine LearningData Science > Computer Algorithms and Programming

    DOI: 10.1002/wcms.1680

    Web of Science

    Scopus

    researchmap

  • ChemTSv2: Democratizing Functional Molecular Design Using de novo Molecule Generator

    Shoichi Ishida, Tanuj Aasawat, Masato Sumita, Michio Katouda, Tatsuya Yoshizawa, Kazuki Yoshizoe, Koji Tsuda, Kei Terayama

    ChemRxiv   2023.2

     More details

  • De novo creation of a naked eye–detectable fluorescent molecule based on quantum chemical computation and machine learning Reviewed

    Masato Sumita, Kei Terayama, Naoya Suzuki, Shinsuke Ishihara, Ryo Tamura, Mandeep K. Chahal, Daniel T. Payne, Kazuki Yoshizoe, Koji Tsuda

    Science Advances   8 ( 10 )   eabj3906   2022.3   ISSN:2375-2548 eISSN:2375-2548

     More details

    Language:Others   Publishing type:Research paper (scientific journal)   Publisher:American Association for the Advancement of Science ({AAAS})  

    Designing fluorescent molecules requires considering multiple interrelated molecular properties, as opposed to properties that straightforwardly correlated with molecular structure, such as light absorption of molecules. In this study, we have used a de novo molecule generator (DNMG) coupled with quantum chemical computation (QC) to develop fluorescent molecules, which are garnering significant attention in various disciplines. Using massive parallel computation (1024 cores, 5 days), the DNMG has produced 3643 candidate molecules. We have selected an unreported molecule and seven reported molecules and synthesized them. Photoluminescence spectrum measurements demonstrated that the DNMG can successfully design fluorescent molecules with 75% accuracy (
    <italic>n</italic>
    = 6/8) and create an unreported molecule that emits fluorescence detectable by the naked eye.

    DOI: 10.1126/sciadv.abj3906

    Web of Science

    Scopus

    PubMed

    researchmap

  • Meta Approach to Data Augmentation Optimization. Reviewed

    Ryuichiro Hataya, Jan Zdenek, Kazuki Yoshizoe, Hideki Nakayama

    Winter Conference on Applications of Computer Vision (WACV)   3535 - 3544   2022.1   ISBN:9781665409155

     More details

    Language:Others   Publishing type:Research paper (other academic)   Publisher:IEEE  

    Data augmentation policies drastically improve the performance of image recognition tasks, especially when the policies are optimized for the target data and tasks. In this paper, we propose to optimize image recognition models and data augmentation policies simultaneously to improve the performance using gradient descent. Unlike prior methods, our approach avoids using proxy tasks or reducing search space, and can directly improve the validation performance. Our method achieves efficient and scalable training by approximating the gradient of policies by implicit gradient with Neumann series approximation. We demonstrate that our approach can improve the performance of various image classification tasks, including fine-grained image recognition, without using dataset-specific hyperparameter tuning.

    DOI: 10.1109/WACV51458.2022.00359

    Scopus

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/wacv/wacv2022.html#HatayaZYN22

  • Meta Approach to Data Augmentation Optimization Reviewed

    美添 一樹

    2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)   1   3535 - 3544   2022

     More details

  • Graph Energy-based Model for Substructure Preserving Molecular Design.

    Ryuichiro Hataya, Hideki Nakayama, Kazuki Yoshizoe

    CoRR   abs/2102.04600   2021.2

     More details

    Language:Others   Publishing type:Research paper (scientific journal)  

  • Machine learning to reveal hidden risk combinations for the trajectory of posttraumatic stress disorder symptoms Reviewed

    Yuta Takahashi, Kazuki Yoshizoe, Masao Ueki, Gen Tamiya, Yu Zhiqian, Yusuke Utsumi, Atsushi Sakuma, Koji Tsuda, Atsushi Hozawa, Ichiro Tsuji, Hiroaki Tomita

    Scientific Reports   10 ( 1 )   2020.12

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    <title>Abstract</title>The nature of the recovery process of posttraumatic stress disorder (PTSD) symptoms is multifactorial. The Massive Parallel Limitless-Arity Multiple-testing Procedure (MP-LAMP), which was developed to detect significant combinational risk factors comprehensively, was utilized to reveal hidden combinational risk factors to explain the long-term trajectory of the PTSD symptoms. In 624 population-based subjects severely affected by the Great East Japan Earthquake, 61 potential risk factors encompassing sociodemographics, lifestyle, and traumatic experiences were analyzed by MP-LAMP regarding combinational associations with the trajectory of PTSD symptoms, as evaluated by the Impact of Event Scale-Revised score after eight years adjusted by the baseline score. The comprehensive combinational analysis detected 56 significant combinational risk factors, including 15 independent variables, although the conventional bivariate analysis between single risk factors and the trajectory detected no significant risk factors. The strongest association was observed with the combination of short resting time, short walking time, unemployment, and evacuation without preparation (adjusted <italic>P</italic> value = 2.2 × 10−4, and raw <italic>P</italic> value = 3.1 × 10−9). Although short resting time had no association with the poor trajectory, it had a significant interaction with short walking time (<italic>P</italic> value = 1.2 × 10−3), which was further strengthened by the other two components (<italic>P</italic> value = 9.7 × 10−5). Likewise, components that were not associated with a poor trajectory in bivariate analysis were included in every observed significant risk combination due to their interactions with other components. Comprehensive combination detection by MP-LAMP is essential for explaining multifactorial psychiatric symptoms by revealing the hidden combinations of risk factors.

    DOI: 10.1038/s41598-020-78966-z

  • HyGN: Hybrid Graph Engine for NUMA Reviewed

    Tanuj Aasawat, Tahsin Reza, Kazuki Yoshizoe, Matei Ripeanu

    2020 IEEE International Conference on Big Data (Big Data)   383 - 390   2020.12

     More details

    Language:English   Publishing type:Research paper (other academic)  

    Modern shared-memory platforms embrace the Non-uniform Memory Access (NUMA) architecture - they have physically distributed, yet cache-coherent shared-memory. This paper explores the feasibility of a shared-memory graph processing engine for NUMA platforms inspired by designs that target zero-sharing platforms. This work exploits the characteristics of two processing modes, synchronous and asynchronous, in the context of the shared-memory NUMA platform. Depending on the algorithm, phase of execution, and graph topology, synchronous and asynchronous modes hold unique advantages over one another. We then explore a hybrid solution that combines synchronous and asynchronous processing within the same graph computation task and harness optimizations therein. An extensive evaluation using graphs with billions of edges and empirical comparisons with several state-of-the-art solutions demonstrate the performance advantages of our design.

    DOI: 10.1109/bigdata50022.2020.9378430

  • On the possibility of short-term traffic prediction during disaster with machine learning approaches: An exploratory analysis Reviewed

    Makoto Chikaraishi, Prateek Garg, Varun Varghese, Kazuki Yoshizoe, Junji Urata, Yasuhiro Shiomi, Ryuki Watanabe

    Transport Policy   98   91 - 104   2020.11

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    Since the cost and time required to finetune parameters in traditional short-term traffic prediction models such as traffic simulators are very high, the prediction models have been developed mainly for managing recurrent congestion, rather than non-recurrent congestion caused, for example, by disaster. Machine learning models are promising candidates for traffic prediction during non-recurrent congestion due to their ability to tune parameters without a-priori knowledge, while their applicability to non-recurrent conditions has rarely been explored. To fill in this gap, this study conducts an exploratory analysis on the applicability of various machine learning models during a transportation network disruption with particular focuses on their ability to predict traffic states and the interpretability of the results. The analysis is conducted by using data obtained during the massive transport network disruption which occurred in Hiroshima in July 2018 due to heavy rain and subsequent landslides. The models tested include random forest, support vector machine, XGBoost, shallow feed-forward neural network, and deep feed-forward neural network. The results indicate that random forest and XGBoost methods produced the best results in terms of prediction accuracy. On the other hand, deep neural network models produce better results in terms of the interpretability of the results, i.e., the results can be logically explained from the perspective of existing traffic flow theory. These findings indicate that the model which produces the best prediction accuracy is not always the best for practical use since it does not mimic the mechanisms of congestion occurrence.

    DOI: 10.1016/j.tranpol.2020.05.023

  • Faster AutoAugment: Learning Augmentation Strategies Using Backpropagation Reviewed

    Ryuichiro Hataya, Jan Zdenek, Kazuki Yoshizoe, Hideki Nakayama

    Computer Vision – ECCV 2020   12370 LNCS   1 - 16   2020.11

     More details

    Language:English   Publishing type:Research paper (other academic)  

    Data augmentation methods are indispensable heuristics to boost the performance of deep neural networks, especially in image recognition tasks. Recently, several studies have shown that augmentation strategies found by search algorithms outperform hand-made strategies. Such methods employ black-box search algorithms over image transformations with continuous or discrete parameters and require a long time to obtain better strategies. In this paper, we propose a differentiable policy search pipeline for data augmentation, which is much faster than previous methods. We introduce approximate gradients for several transformation operations with discrete parameters as well as a differentiable mechanism for selecting operations. As the objective of training, we minimize the distance between the distributions of augmented and original data, which can be differentiated. We show that our method, Faster AutoAugment, achieves significantly faster searching than prior methods without a performance drop.

    DOI: 10.1007/978-3-030-58595-2_1

  • CompRet: a comprehensive recommendation framework for chemical synthesis planning with algorithmic enumeration Reviewed

    Ryosuke Shibukawa, Shoichi Ishida, Kazuki Yoshizoe, Kunihiro Wasa, Kiyosei Takasu, Yasushi Okuno, Kei Terayama, Koji Tsuda

    Journal of Cheminformatics   12 ( 1 )   52   2020.9

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    <title>Abstract</title>In computer-assisted synthesis planning (CASP) programs, providing as many chemical synthetic routes as possible is essential for considering optimal and alternative routes in a chemical reaction network. As the majority of CASP programs have been designed to provide one or a few optimal routes, it is likely that the desired one will not be included. To avoid this, an exact algorithm that lists possible synthetic routes within the chemical reaction network is required, alongside a recommendation of synthetic routes that meet specified criteria based on the chemist’s objectives. Herein, we propose a chemical-reaction-network-based synthetic route recommendation framework called “CompRet” with a mathematically guaranteed enumeration algorithm. In a preliminary experiment, CompRet was shown to successfully provide alternative routes for a known antihistaminic drug, cetirizine. CompRet is expected to promote desirable enumeration-based chemical synthesis searches and aid the development of an interactive CASP framework for chemists.

    DOI: 10.1186/s13321-020-00452-5

  • NMR-TS: de novo molecule identification from NMR spectra Reviewed

    Jinzhe Zhang, Kei Terayama, Masato Sumita, Kazuki Yoshizoe, Kengo Ito, Jun Kikuchi, Koji Tsuda

    Science and Technology of Advanced Materials   21 ( 1 )   552 - 561   2020.7

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.1080/14686996.2020.1793382

  • Faster AutoAugment:誤差逆伝播法によるデータ拡張の学習

    幡谷龍一郎, 幡谷龍一郎, ZDENEK Jan, 美添一樹, 中山英樹

    人工知能学会全国大会(Web)   34th   1 - 16   2020.6

     More details

    Language:Others   Publishing type:Research paper (other academic)  

    Faster AutoAugment: Learning Augmentation Strategies Using Backpropagation.

    DOI: 10.1007/978-3-030-58595-2_1

  • Meta Approach to Data Augmentation Optimization.

    Ryuichiro Hataya, Jan Zdenek, Kazuki Yoshizoe, Hideki Nakayama

    CoRR   abs/2006.07965   2020.6

     More details

    Language:Others   Publishing type:Research paper (scientific journal)  

  • MP-LAMP: parallel detection of statistically significant multi-loci markers on cloud platforms Reviewed

    Kazuki Yoshizoe, Aika Terada, Koji Tsuda

    Bioinformatics   34 ( 17 )   3047 - 3049   2018.9

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.1093/bioinformatics/bty219

  • Krawczyk-Hansenによる精度保証つき大域的最適化法の局所解を用いた高速化 (情報論的学習理論と機械学習)

    高田 浩彰, 美添 一樹, 石井 大輔, 津田 宏治

    電子情報通信学会技術研究報告 = IEICE technical report : 信学技報   117 ( 475 )   63 - 70   2018.3

     More details

    Language:Japanese  

    Using local minima to accelerate Krawczyk-Hansen global optimization

  • An Extended GLB Library for Optimization Problems Reviewed

    Shota Izumi, Daisuke Ishii, Kazuki Yoshizoe

    2018.1

     More details

    Language:English   Publishing type:Research paper (other academic)  

    An Extended GLB Library for Optimization Problems

  • ChemTS: an efficient python library for de novo molecular generation Reviewed

    Xiufeng Yang, Jinzhe Zhang, Kazuki Yoshizoe, Kei Terayama, Koji Tsuda

    Science and Technology of Advanced Materials   18 ( 1 )   972 - 976   2017.12

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    Automatic design of organic materials requires black-box optimization in a vast chemical space. In conventional molecular design algorithms, a molecule is built as a combination of predetermined fragments. Recently, deep neural network models such as variational autoencoders and recurrent neural networks (RNNs) are shown to be effective in de novo design of molecules without any predetermined fragments. This paper presents a novel Python library ChemTS that explores the chemical space by combining Monte Carlo tree search and an RNN. In a benchmarking problem of optimizing the octanol-water partition coefficient and synthesizability, our algorithm showed superior efficiency in finding high-scoring molecules. ChemTS is available at https://github.com/tsudalab/ChemTS.[GRAPHICS].

    DOI: 10.1080/14686996.2017.1401424

  • RNA inverse folding using Monte Carlo tree search Reviewed

    Xiufeng Yang, Kazuki Yoshizoe, Akito Taneda, Koji Tsuda

    BMC Bioinformatics   18 ( 1 )   2017.12

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    Background: Artificially synthesized RNA molecules provide important ways for creating a variety of novel functional molecules. State-of-the-art RNA inverse folding algorithms can design simple and short RNA sequences of specific GC content, that fold into the target RNA structure. However, their performance is not satisfactory in complicated cases.Result: We present a new inverse folding algorithm called MCTS-RNA, which uses Monte Carlo tree search (MCTS), a technique that has shown exceptional performance in Computer Go recently, to represent and discover the essential part of the sequence space. To obtain high accuracy, initial sequences generated by MCTS are further improved by a series of local updates. Our algorithm has an ability to control the GC content precisely and can deal with pseudoknot structures. Using common benchmark datasets for evaluation, MCTS-RNA showed a lot of promise as a standard method of RNA inverse folding.Conclusion: MCTS-RNA is available at https://github.com/tsudalab/MCTS-RNA.

    DOI: 10.1186/s12859-017-1882-7

  • MDTS: automatic complex materials design using Monte Carlo tree search Reviewed

    Thaer M. Dieb, Shenghong Ju, Kazuki Yoshizoe, Zhufeng Hou, Junichiro Shiomi, Koji Tsuda

    Science and Technology of Advanced Materials   18 ( 1 )   498 - 503   2017.12

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.1080/14686996.2017.1344083

  • Redesigning pattern mining algorithms for supercomputers.

    Kazuki Yoshizoe, Aika Terada, Koji Tsuda

    CoRR   abs/1510.07787   2015.10

     More details

    Language:Others   Publishing type:Research paper (scientific journal)  

  • モンテカルロ木探索を統合したプレイアウト方策の最適化

    渡辺 順哉, 美添 一樹, 金子 知適

    ゲームプログラミングワークショップ2015論文集   ( 2015 )   5 - 11   2015.10

     More details

    Language:Japanese  

    Optimization of Playout Policy Integrated with Monte-Carlo Tree Search

  • Scalable parallel numerical constraint solver using global load balancing Reviewed

    Daisuke Ishii, Kazuki Yoshizoe, Toyotaro Suzumura

    Proceedings of the ACM SIGPLAN Workshop on X10   33 - 38   2015.6

     More details

    Language:English   Publishing type:Research paper (other academic)  

    We present a scalable parallel solver for numerical constraint satisfaction problems (NCSPs). Our parallelization scheme consists of homogeneous worker solvers, each of which runs on an available core and communicates with others via the global load balancing (GLB) method. The search tree of the branch and prune algorithm is split and distributed through the two phases of GLB: a random workload stealing phase and a workload distribution and termination phase based on a hyper-cube-shaped graph called lifeline. The parallel solver is simply implemented with X10 that provides an implementation of GLB as a library. In experiments, NCSPs from the literature were solved and attained up to 516-fold speedup using 600 cores of the TSUBAME2.5 supercomputer. Optimal GLB configurations are analyzed.

    DOI: 10.1145/2771774.2771776

  • Scalable Parallel Numerical CSP Solver Reviewed

    Daisuke Ishii, Kazuki Yoshizoe, Toyotaro Suzumura

    Lecture Notes in Computer Science   398 - 406   2014.9

     More details

    Language:English  

    DOI: 10.1007/978-3-319-10428-7_30

  • 進化計算とUCTによるMarioを人間らしくプレイするAI

    中野 雄基, 美添 一樹, 脇田 建

    ゲームプログラミングワークショップ2013論文集   81 - 88   2013.11

     More details

    Language:Japanese  

    Humanlike AI for Mario Bros. Based on Evolutionary Computation and UCT

  • Accelerated UCT and Its Application to Two-Player Games Reviewed

    Junichi Hashimoto, Akihiro Kishimoto, Kazuki Yoshizoe, Kokolo Ikeda

    Lecture Notes in Computer Science   1 - 12   2012.1

     More details

    Language:English  

    DOI: 10.1007/978-3-642-31866-5_1

  • Triple Line-Based Playout for Go - An Accelerator for Monte Carlo Go Reviewed

    Kenichi Koizumi, Mary Inaba, Kei Hiraki, Yasuo Ishii, Takefumi Miyoshi, Kazuki Yoshizoe

    2009 International Conference on Reconfigurable Computing and FPGAs   161 - 166   2009.12

     More details

    Language:English   Publishing type:Research paper (other academic)  

    After a computer named "Deep Blue" defeated the world chess champion Garry Kasparov in 1997, researchers studying computer board games focused their attention on the game "Go." Go is known to be more difficult for computers to play than chess or shogi because (1) the search space for Go is much larger, (2) it is difficult to define an appropriate evaluation function of position, and (3) a position sometimes changes globally in just one move. Recently, a new meth ad called Monte Carlo Go has been developed, which involves performing Monte Carlo simulations to evaluate a position. Monte Carlo Go increases the strength of the Computer-Go program. For Monte Carlo Go, the strength fully depends on the number of simulations. Several attempts were made to accelerate simulations, e.g., by the use of cluster systems and FPGAs. The cluster system yields good results, but it is a very expensive system. On the other hand, acceleration using an FPGA was not so easy because the usage of FPGA resources tends to be high. Previously, FPGA acceleration was feasible for smaller board such as a board with a 9 x 9 grid, while it was not feasible for the standard board with a 19 x 19 grid. In this paper, we propose triple line-based playout for Go (TLPG), a hardware algorithm for generating simulations using an FPGA. By reproducing global information redundantly, TLPG enables the generation of simulations only using local operations; this helps realize compact implementations of hardware logic, and thus, TLPG can handle both 9 x 9 and 19 x 19 grid Go boards. We implement TLPG on Xilinx Virtex-5 (XC5VFX70T-IFF1136) and evaluate it. TLPG can perform 40,649 playouts per second for a 9 x 9 grid Go board and 4,668 playouts per second for a 19 x 19 grid Go board.

    DOI: 10.1109/reconfig.2009.75

  • A study on security evaluation methodology for image-based biometrics authentication systems Reviewed

    Yasuhiro Tanabe, Kazuki Yoshizoe, Hideki Imai

    2009 IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems   2009.9

     More details

    Language:English   Publishing type:Research paper (other academic)  

    We propose here a security evaluation methodology of image-based biometrics authentication systems against wolf attacks. A wolf attack is an attempt to impersonate a victim by feeding wolves into the system to be attacked. The wolf is input data that can be falsely accepted as a match with multiple templates. To create a secure system, we must evaluate the possibility of wolf attacks. Existing studies have relied on theoretical analysis of algorithms carried out by human beings, which is only effective if theoretical analysis is possible. Therefore, we propose a more generic approach based on a search to assist the developers of matching algorithms. We searched for wolves by using a recently developed algorithm called Monte-Carlo Tree Search (MCTS). We succeeded in detecting wolves in a matching algorithm, which appears promising considering that this is the first trial for this kind of approach. ©2009 IEEE.

    DOI: 10.1109/btas.2009.5339016

  • 分岐因子が一様な探索空間のためのAND-OR木探索アルゴリズム

    美添 一樹

    博士論文. 東京大学大学院情報理工学系研究科   2009.2

     More details

    Language:English  

    AND-OR Tree Search Algorithms for Domains with Uniform Branching Factors

  • A New Proof-Number Calculation Technique for Proof-Number Search Reviewed

    Kazuki Yoshizoe

    Computers and Games   135 - 145   2008.9

     More details

    Language:English  

    DOI: 10.1007/978-3-540-87608-3_13

  • A Privacy Protection Scheme for a Scalable Control Method in Context-Dependent Services Reviewed

    Rei Yoshida, Rie Shigetomi, Kazuki Yoshizoe, Akira Otsuka, Hideki Imai

    WEWoRC 2007: Research in Cryptology   1 - 12   2008.7

     More details

    Language:English  

    DOI: 10.1007/978-3-540-88353-1_1

  • 指静脈認証システムにおけるセキュリティ評価手法の提案.

    田辺 康宏, 美添 一樹, 今井 秀樹

    2008年 暗号と情報セキュリティシンポジウム(SCIS2008)   3B-42   2008.1

     More details

    Language:Japanese  

  • 証明数と反証数を用いたλ探索 Reviewed

    副田俊介, 美添一樹, 岸本章宏, 金子知適, 田中哲朗, マーティンミュラー

    情報処理学会論文誌   48 ( 11 )   3455 - 3462   2007.11

     More details

    Language:Japanese  

    λ Search Based on Proof and Disproof Numbers
    We present the df-pn λ search algorithm that combines threats with proof and disproof numbers. λ search is a promising method based on threats. Df-pn is an efficient algorithm that employs the notion of proof and disproof numbers. However, λ search uses neither proof nor disproof numbers, whereas df-pn incorporates no information on threat levels. Integrating threats with proof and disproof numbers is a natural extension to further enhance the search performance. We introduce pseudo-nodes for various threat levels at each node, to represent a node searched with a specific threat level. Then the proof and disproof numbers of the original node are defined using pseudo-nodes, which provides a model that can be searched with df-pn. We compared df-pn λ with df-pn on games with different properties. The results showed that df-pn λ is better than df-pn in Shogi and Go.

  • 証明数を用いたλ探索の効率的な実装

    副田俊介, 美添一樹, 田中哲朗

    ゲームプログラミングワークショップ2006論文集   2006.11

     More details

    Language:Others  

    Efficient Implementation of the Lambda Search based on Proof Numbers

  • Monte carlo go has a way to go Reviewed

    Haruhiro Yoshimoto, Kazuki Yoshizoe, Tomoyuki Kaneko, Akihiro Kishimoto, Kenjiro Taura

    Twenty-First National Conference on Artificial Intelligence (AAAI-06)   1070 - 1075   2006.7

     More details

    Language:English   Publishing type:Research paper (other academic)  

  • A search algorithm for finding multi purpose moves in sub problems of Go

    Kazuki Yoshizoe

    10th Game Programming Workshop (GPW05)   10   76 - 83   2005.11

     More details

    Language:English  

  • HSDPAにおけるピーク抑圧閾値と誤り率の関係についての一検討

    美添 一樹, 齋藤 直之, 岩松 隆則

    電子情報通信学会2004ソサイエティ大会論文集   2004   B-5-31   2004.9

     More details

    Language:Japanese  

  • Speculative Parallel Execution on JVM Reviewed

    YOSHIZOE K.

    1st UK Workshop on Java for High Performance Network Computing   1998.9

     More details

    Language:English   Publishing type:Research paper (other academic)  

    Speculative Parallel Execution on JVM

▼display all

Books

  • コンピュータ囲碁 : モンテカルロ法の理論と実践

    松原, 仁, 美添, 一樹, 山下, 宏

    共立出版  2012.11 

     More details

    Responsible for pages:総ページ数:xi, 222p   Language:Japanese  

Presentations

  • スケーラブルな並列探索による最適化問題の求解

    泉 翔太, 石井 大輔, 美添 一樹

    第167回HPC研究会  2018.12 

     More details

    Event date: 2018.12

    Language:Others  

    Country:Other  

  • スケーラブルな並列探索による最適化問題の求解

    泉 翔太, 石井 大輔, 美添 一樹

    第167回HPC研究会  2018.12 

     More details

    Event date: 2018.12

    researchmap

  • Krawczyk-Hansenによる精度保証つき大域的最適化法の局所解を用いた高速化 (情報論的学習理論と機械学習)

    高田 浩彰, 美添 一樹, 石井 大輔, 津田 宏治

    電子情報通信学会技術研究報告 = IEICE technical report : 信学技報  2018.3 

     More details

    Event date: 2018.3

    Language:Japanese  

    Country:Other  

    Using local minima to accelerate Krawczyk-Hansen global optimization

  • Using local minima to accelerate Krawczyk-Hansen global optimization

    高田 浩彰, 美添 一樹, 石井 大輔, 津田 宏治

    電子情報通信学会技術研究報告 = IEICE technical report : 信学技報  2018.3  電子情報通信学会

     More details

    Event date: 2018.3

    Language:Japanese  

    researchmap

  • X10GLBライブラリの最適化問題のための拡張

    泉翔太, 石井大輔, 美添一樹

    電気関係学会北陸支部連合大会講演論文集  2017.9 

     More details

    Event date: 2017.9

    Language:Others  

    Country:Other  

  • X10GLBライブラリの最適化問題のための拡張

    泉翔太, 石井大輔, 美添一樹

    電気関係学会北陸支部連合大会講演論文集  2017.9 

     More details

    Event date: 2017.9

    researchmap

  • 数値制約ソルバーのスケーラブルな並列化

    石井 大輔, 美添 一樹, 鈴村 豊太郎

    日本ソフトウェア科学会大会論文集  2015.9 

     More details

    Event date: 2015.9

    Language:Japanese  

    Country:Other  

    Scalable Parallelization of a Numerical Constraint Solver

  • Scalable Parallelization of a Numerical Constraint Solver

    石井 大輔, 美添 一樹, 鈴村 豊太郎

    日本ソフトウェア科学会大会論文集  2015.9  [日本ソフトウェア科学会]

     More details

    Event date: 2015.9

    Language:Japanese  

    researchmap

  • 分散並列モンテカルロ木探索フレームワークの提案

    美添 一樹, 石川 裕

    情報処理学会研究報告. [ハイパフォーマンスコンピューティング]  2010.7 

     More details

    Event date: 2010.7

    Language:Japanese  

    Country:Other  

    A Proposal for Distributed Parallel Monte Carlo Tree Search Framework
    We propose to implement a framework for Monte Carlo Tree Search (MCTS) on distributed parallel systems. The objective is to facilitate the parallelization of existing sequential Monte Carlo Tree Search programs. It is expected to be effective for highly parallel distributed systems by using Transposition table Driven Scheduling as the parallelization technique for the framework.

  • A Proposal for Distributed Parallel Monte Carlo Tree Search Framework

    Kazuki Yoshizoe, Yutaka Ishikawa

    IPSJ SIG Notes  2010.7  Information Processing Society of Japan (IPSJ)

     More details

    Event date: 2010.7

    Language:Japanese  

    We propose to implement a framework for Monte Carlo Tree Search (MCTS) on distributed parallel systems. The objective is to facilitate the parallelization of existing sequential Monte Carlo Tree Search programs. It is expected to be effective for highly parallel distributed systems by using Transposition table Driven Scheduling as the parallelization technique for the framework.

    researchmap

  • モンテカルロ法によるゲームAIの可能性. Invited

    美添一樹

    CESA Developer's Conference (CEDEC2009)  2009.9 

     More details

    Event date: 2009.9

    Language:Japanese  

    Country:Other  

  • FPGA基板を用いたモンテカルロ碁の高速化(アクセラレーションと回路設計,2009年並列/分散/協調処理に関する『仙台』サマー・ワークショップ(SWoPP仙台2009))

    小泉賢一, 石井康雄, 美添一樹, 三好健文, 菅原豊, 稲葉真理, 平木敬

    電子情報通信学会技術研究報告. CPSY, コンピュータシステム  2009.7 

     More details

    Event date: 2009.7

    Language:Japanese  

    Country:Other  

    Acceleration of Monte-Carlo Go by FPGA-based Hardware
    In the monte-carlo simulation of Go, it takes time to run playouts. There were attempts of accelerating by implementing circuits for playout on FPGA, but it is difficult to realize high-speed playouts because of high utilization of resources in a FPGA. In this paper, we propose an algorithm, Triple Line-based Playout for Go (TLPG) to accelerate playouts for the monte-carlo tree search for computer-go game. We implemented the playout logics on FPGA for 9x9 and 19x19 boards. With the optimizations, We achieved 13104playouts/sec in 9x9 and 2055playouts/sec in 19x19 board in simulation. By making games with GNU Go on a host Computer, We evaluation the playouts of TLPG.

  • Acceleration of Monte-Carlo Go by FPGA-based Hardware

    KOIZUMI Kenichi, ISHII Yasuo, YOSHIZOE Kazuki, MIYOSHI Takefumi, SUGAWARA Yutaka, INABA Mary, HIRAKI Kei

    IEICE technical report. Computer systems  2009.7  The Institute of Electronics, Information and Communication Engineers

     More details

    Event date: 2009.7

    Language:Japanese  

    In the monte-carlo simulation of Go, it takes time to run playouts. There were attempts of accelerating by implementing circuits for playout on FPGA, but it is difficult to realize high-speed playouts because of high utilization of resources in a FPGA. In this paper, we propose an algorithm, Triple Line-based Playout for Go (TLPG) to accelerate playouts for the monte-carlo tree search for computer-go game. We implemented the playout logics on FPGA for 9x9 and 19x19 boards. With the optimizations, We achieved 13104playouts/sec in 9x9 and 2055playouts/sec in 19x19 board in simulation. By making games with GNU Go on a host Computer, We evaluation the playouts of TLPG.

    researchmap

  • 動的候補手拡大を用いたDf-pnアルゴリズム

    美添一樹

    ゲームプログラミングワークショップ2007論文集  2007.11 

     More details

    Event date: 2007.11

    Language:Japanese  

    Country:Other  

    Df-pn Algorithm with Dynamic Widening

  • Df-pn Algorithm with Dynamic Widening

    美添一樹

    ゲームプログラミングワークショップ2007論文集  2007.11 

     More details

    Event date: 2007.11

    Language:Japanese  

    researchmap

  • DFUCTの囲碁への応用について

    吉本晴洋, 岸本章宏, 金子知適, 美添一樹

    情報処理学会シンポジウム論文集  2007.11 

     More details

    Event date: 2007.11

    Language:Others  

    Country:Other  

    Depth-First UCT and Its Application to Go

  • Randamized Response Techniqueのコンテキスト依存型サービス適用検討

    吉田伶, 繁富利恵, 美添一樹, 今井秀樹, 今井秀樹

    情報理論とその応用シンポジウム予稿集  2007.11 

     More details

    Event date: 2007.11

    Language:Others  

    Country:Other  

  • Randamized Response Techniqueのコンテキスト依存型サービス適用検討

    吉田伶, 繁富利恵, 美添一樹, 今井秀樹, 今井秀樹

    情報理論とその応用シンポジウム予稿集  2007.11 

     More details

    Event date: 2007.11

    researchmap

  • Depth-First UCT and Its Application to Go

    吉本晴洋, 岸本章宏, 金子知適, 美添一樹

    情報処理学会シンポジウム論文集  2007.11 

     More details

    Event date: 2007.11

    researchmap

  • 囲碁の部分問題における両利きの探索

    美添一樹, 今井浩

    情報処理学会研究報告. GI, [ゲーム情報学]  2005.9 

     More details

    Event date: 2005.9

    Language:English  

    Country:Other  

    Searching for Double Threats in Subproblems of the Game of Go
    It is difficult to make a fast and accurate evaluation function for the whole board in the Game of Go. Therefore sub-goal directed search is used widely among Go playing programs. One problem of sub-goal directed search is dependencies between sub-goals. There are several researches which aim to resolve the dependencies by obtaining the area which involves with the result of subgoals. An idea called relevancy zone is being used in some researches. In this paper, we introduce an algorithm which search for an area which would improve relevancy zone. The intersection of two such areas will be the candidate for double threat.

  • ループ並列投機実行のJava仮想マシンの適用

    美添一樹, 松本尚, 平木敬

    情報処理学会研究報告. [ハイパフォーマンスコンピューティング]  1998.8 

     More details

    Event date: 1998.8

    Language:Japanese  

    Country:Other  

    Implementing Parallel Speculative Execution of Loops on JVM
    There have been several proposals about hardware speculative executions, in a larger gran-ularity than instruction level parallelism, by partitioning the target program into blocks.We have applied speculative execution onto Java Virtual Machine. We implemented it on a shared memory machine. The target for speculative execution is limited to loops. We measured speedups for simple loops and found that it is possible to gain speedups for loops which contains more than 10000 instructions by an interpreter Java Virtual Machine.

  • Implementing Parallel Speculative Execution of Loops on JVM

    YOSHIZOE KAZUKI, MATSUMOTO TAKASHI, HIRAKI KEI

    IPSJ SIG Notes  1998.8  Information Processing Society of Japan (IPSJ)

     More details

    Event date: 1998.8

    Language:Japanese  

    There have been several proposals about hardware speculative executions, in a larger gran-ularity than instruction level parallelism, by partitioning the target program into blocks.We have applied speculative execution onto Java Virtual Machine. We implemented it on a shared memory machine. The target for speculative execution is limited to loops. We measured speedups for simple loops and found that it is possible to gain speedups for loops which contains more than 10000 instructions by an interpreter Java Virtual Machine.

    researchmap

  • ループ並列投機実行のJava仮想マシンの適用

    美添一樹, 松本尚, 平木敬

    情報処理学会研究報告. [ハイパフォーマンスコンピューティング]  1998.8 

     More details

    Language:Japanese  

    Country:Japan  

    Implementing Parallel Speculative Execution of Loops on JVM
    There have been several proposals about hardware speculative executions, in a larger gran-ularity than instruction level parallelism, by partitioning the target program into blocks.We have applied speculative execution onto Java Virtual Machine. We implemented it on a shared memory machine. The target for speculative execution is limited to loops. We measured speedups for simple loops and found that it is possible to gain speedups for loops which contains more than 10000 instructions by an interpreter Java Virtual Machine.

  • 囲碁の部分問題における両利きの探索(Session 3)

    美添一樹, 今井浩

    情報処理学会研究報告. GI, [ゲーム情報学]  2005.9 

     More details

    Language:English  

    Country:Japan  

    Searching for Double Threats in Subproblems of the Game of Go(Session 3)
    It is difficult to make a fast and accurate evaluation function for the whole board in the Game of Go. Therefore sub-goal directed search is used widely among Go playing programs. One problem of sub-goal directed search is dependencies between sub-goals. There are several researches which aim to resolve the dependencies by obtaining the area which involves with the result of subgoals. An idea called relevancy zone is being used in some researches. In this paper, we introduce an algorithm which search for an area which would improve relevancy zone. The intersection of two such areas will be the candidate for double threat.

  • 動的候補手拡大を用いたDf-pnアルゴリズム

    美添一樹

    ゲームプログラミングワークショップ2007論文集  2007.11 

     More details

    Language:Japanese  

    Country:Japan  

    Df-pn Algorithm with Dynamic Widening

  • Randamized Response Techniqueのコンテキスト依存型サービス適用検討

    吉田伶, 繁富利恵, 美添一樹, 今井秀樹, 今井秀樹

    情報理論とその応用シンポジウム予稿集  2007.11 

     More details

    Language:Others  

    Country:Japan  

  • DFUCTの囲碁への応用について

    吉本晴洋, 岸本章宏, 金子知適, 美添一樹

    情報処理学会シンポジウム論文集  2007.11 

     More details

    Language:Others  

    Country:Japan  

    Depth-First UCT and Its Application to Go

  • コンピュータ囲碁におけるモンテカルロ法. Invited

    美添一樹

    エンターテイメントと認知科学研究ステーション 第5回講演会 (電気通信大学)  2008.6 

     More details

    Language:Japanese  

    Country:Other  

  • 囲碁AIにおける革命「モンテカルロ木探索」とは何か? Invited

    美添一樹

    DigraJapan(日本デジタルゲーム学会) 公開講座08年11月, 東京大学  2008.11 

     More details

    Language:Japanese  

    Country:Other  

  • FPGA基板を用いたモンテカルロ碁の高速化(アクセラレーションと回路設計,2009年並列/分散/協調処理に関する『仙台』サマー・ワークショップ(SWoPP仙台2009))

    小泉賢一, 石井康雄, 美添一樹, 三好健文, 菅原豊, 稲葉真理, 平木敬

    電子情報通信学会技術研究報告. CPSY, コンピュータシステム  2009.7 

     More details

    Language:Japanese  

    Country:Japan  

    Acceleration of Monte-Carlo Go by FPGA-based Hardware
    In the monte-carlo simulation of Go, it takes time to run playouts. There were attempts of accelerating by implementing circuits for playout on FPGA, but it is difficult to realize high-speed playouts because of high utilization of resources in a FPGA. In this paper, we propose an algorithm, Triple Line-based Playout for Go (TLPG) to accelerate playouts for the monte-carlo tree search for computer-go game. We implemented the playout logics on FPGA for 9x9 and 19x19 boards. With the optimizations, We achieved 13104playouts/sec in 9x9 and 2055playouts/sec in 19x19 board in simulation. By making games with GNU Go on a host Computer, We evaluation the playouts of TLPG.

  • 分散並列モンテカルロ木探索フレームワークの提案

    美添 一樹, 石川 裕

    情報処理学会研究報告. [ハイパフォーマンスコンピューティング]  2010.7 

     More details

    Language:Japanese  

    Country:Japan  

    A Proposal for Distributed Parallel Monte Carlo Tree Search Framework
    We propose to implement a framework for Monte Carlo Tree Search (MCTS) on distributed parallel systems. The objective is to facilitate the parallelization of existing sequential Monte Carlo Tree Search programs. It is expected to be effective for highly parallel distributed systems by using Transposition table Driven Scheduling as the parallelization technique for the framework.

  • Accelerated UCT and Its Application to Two-Player Games International conference

    Junichi Hashimoto, Akihiro Kishimoto, Kazuki Yoshizoe, Kokolo Ikeda

    Advances in Computer Games 13  2011.11 

     More details

    Language:English  

    Country:Other  

  • 数値制約ソルバーのスケーラブルな並列化

    石井 大輔, 美添 一樹, 鈴村 豊太郎

    日本ソフトウェア科学会大会論文集  2015.9 

     More details

    Language:Japanese  

    Country:Japan  

    Scalable Parallelization of a Numerical Constraint Solver

  • モンテカルロ木探索を統合したプレイアウト方策の最適化

    渡辺 順哉, 美添 一樹, 金子 知適

    ゲームプログラミングワークショップ2015論文集  2015.10 

     More details

    Language:Japanese  

    Country:Japan  

    Optimization of Playout Policy Integrated with Monte-Carlo Tree Search

  • An Extended GLB Library for Optimization Problems

    Shota Izumi, Daisuke Ishii, Kazuki Yoshizoe

    2018.1 

     More details

    Language:Others  

    Country:Japan  

    An Extended GLB Library for Optimization Problems

  • Krawczyk-Hansenによる精度保証つき大域的最適化法の局所解を用いた高速化 (情報論的学習理論と機械学習)

    高田 浩彰, 美添 一樹, 石井 大輔, 津田 宏治

    電子情報通信学会技術研究報告 = IEICE technical report : 信学技報  2018.3 

     More details

    Language:Japanese  

    Country:Japan  

    Using local minima to accelerate Krawczyk-Hansen global optimization

  • スケーラブルな並列探索による最適化問題の求解

    泉 翔太, 石井 大輔, 美添 一樹

    第167回HPC研究会  2018.12 

     More details

    Language:Others  

    Country:Japan  

  • 東日本大震災被災者における外傷後ストレス障害症状の変化を予測する因子に関する,機械学習を用いた組み合わせの検討

    高橋雄太, 高橋雄太, 高橋雄太, 美添一樹, 植木優夫, 植木優夫, 田宮元, 田宮元, 富田博秋, 富田博秋, 富田博秋

    日本生物学的精神医学会  2019.6 

     More details

    Language:Others  

    Country:Japan  

▼display all

MISC

  • 5分で分かる! ? 有名論文ナナメ読み:Silver, D. et al. : Mastering the Game of Go without Human Knowledge

    美添 一樹

    情報処理   2018.7

     More details

    Language:Japanese  

    Skimming a Famous Paper in Five Minutes

  • 人工知能をさまざまな分野で活用する 探索アルゴリズムによりディープラーニングをさらに高性能に

    美添一樹

    広報誌RIKEN   2018.6

     More details

    Language:Others  

  • 近年のゲームAIアルゴリズムの進歩について

    美添一樹

    人工知能学会人工知能基本問題研究会資料   2017.8

     More details

    Language:Others  

  • モンテカルロ木探索とその組合せ最適化問題への応用

    美添一樹

    電子情報通信学会大会講演論文集   2017.3

     More details

    Language:Others  

  • Computer Go

    Kazuki Yoshizoe, Martin Müller

    Encyclopedia of Computer Graphics and Games   2016.2

     More details

    Language:Others  

    DOI: 10.1007/978-3-319-08234-9_20-1

  • 超大規模なグラフ構造の効率的な処理技術 (小特集 「フカシギの数え方」から広がるアルゴリズムの理工学 : 二分決定グラフによる離散構造処理と広がる応用分野)

    戸田 貴久, 竹内 聖悟, 美添 一樹

    電子情報通信学会誌 = The journal of the Institute of Electronics, Information and Communication Engineers   2014.12

     More details

    Language:Japanese  

    Efficient Processing Techniques for Very Large-scale Graph Structure

  • 連載:理学のキーワード : 第30回

    美添 一樹, 藤森 淳, 今村 剛, 田中 秀幸, 野中 勝, 吉川 一朗

    東京大学理学系研究科・理学部ニュース   2011.3

     More details

    Language:Japanese  

    「ゲーム木検索」/「スピントロニクス」/「金星のスーパーローテーション」/「グラフェン」/「獲得免疫と自然免疫」/(あとがき)

  • コンピュータ囲碁の飛躍の背景--モンテカルロ木探索

    美添 一樹, 村松 正和

    数学セミナー   2010.10

     More details

    Language:Japanese  

  • 分岐因子が一様な探索空間のためのAND-OR木探索アルゴリズム(基礎・理論,<特集>人工知能分野における博士論文)

    美添一樹

    人工知能学会誌   2010.1

     More details

    Language:Japanese  

    AND-OR Tree Search Algorithms for Domains with Uniform Branching Factors(Foundations of AI,<Special Issue>Doctorial Theses on Aritifical Intelligence)

    DOI: 10.11517/jjsai.25.1_148

  • モンテカルロ木探索 : コンピュータ囲碁に革命を起こした新手法

    美添一樹

    情報処理   2008.6

     More details

    Language:Japanese  

    Monte-Carlo Tree Search : A Revolutionary Algorithm Developed for Computer Go

▼display all

Industrial property rights

Patent   Number of applications: 1   Number of registrations: 1
Utility model   Number of applications: 0   Number of registrations: 0
Design   Number of applications: 0   Number of registrations: 0
Trademark   Number of applications: 0   Number of registrations: 0

Professional Memberships

▼display all

Committee Memberships

  • 情報処理学会 ゲーム情報学研究会   Organizer   Domestic

    2018.4 - 2022.3   

  • 情報処理学会 ゲーム情報学研究会   幹事   Domestic

    2018.4 - 2022.3   

Academic Activities

  • Program Committee International contribution

    Thirty-fifth Conference on Neural Information Processing Systems  ( Virtual-only (オンライン) その他 ) 2021.12

     More details

    Type:Competition, symposium, etc. 

    Number of participants:15,000

  • Program Committee (reviewer) International contribution

    Thirty-eighth International Conference on Machine Learning  ( Virtual Only (オンライン) その他 ) 2021.7

     More details

    Type:Competition, symposium, etc. 

Research Projects

  • 社会を志向した革新的アルゴリズムの実装

    2021.10 - 2025.3

      More details

    Authorship:Coinvestigator(s) 

  • 社会変革の源泉となる革新的アルゴリズム基盤の創出と体系化

    Grant number:20H05963  2021 - 2025

    Japan Society for the Promotion of Science・Ministry of Education, Culture, Sports, Science and Technology  Grants-in-Aid for Scientific Research  Grant-in-Aid for Transformative Research Areas (A)

    安田 宜仁, 鍋島 英知, 有村 博紀, 井上 武, 美添 一樹, 西野 正彬

      More details

    Authorship:Coinvestigator(s)  Grant type:Scientific research funding

    本研究班では、領域内他研究班で得られる成果を「アルゴリズム基盤」という共通の場に実装することにより、応用研究者が個別のアルゴリズムを探索・検討せずとも最先端の革新的アルゴリズム群を利用できる仕組みを構築する。
    こうした仕組みがうまく機能するためには、アルゴリズムが持つ数学的な構造は維持しつつ、かつ、応用研究者は問題を容易に記述できるようなアルゴリズムと応用の「良いつなぎ方」が求められる。そこで本研究班では、「良いつなぎ方」の探索を中心的な課題と捉え、実応用問題を用いたケーススタディを通じてそのあり方を探る。

    CiNii Research

  • High-Performance Optimization Algorithm based on Machine Learning and Search

    Grant number:23K20387  2020.4 - 2025.3

    Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (B)

    美添 一樹

      More details

    Grant type:Scientific research funding

    機械学習手法の発展によって従来数値で表せなかった様々なデータが計算機で効率良く処理できるようになっている。進歩した機械学習モデルを用いて最適化や探索を行うことによって、さらに多くの分野で実問題を解くことが可能となりつつある。機械学習手法単独での研究は盛んに行われているが、探索アルゴリズムと組み合わせた場合の研究は世界的に不足している。本研究では機械学習と探索アルゴリズムを組み合わせて、汎用性が高く高性能な最適化手法を開発することにより、多くの実問題に適用可能なソルバーの実現へ繋げることを目指す。

    CiNii Research

  • 機械学習と探索の協調による高性能最適化アルゴリズム

    2020.4

      More details

    Authorship:Principal investigator 

    機械学習を他のアルゴリズムを併用することにより、機械学習単独で解くことが難しい複雑な意思決定問題を解くことが可能となっている。しかし高い能力を持つ有望な手法でありながら、適用対象は限られている。
    一つの理由は、探索アルゴリズムは機械学習と比較して利用が難しいことである。これは実装面の難しさと汎用性の二つの理由がある。もう一つは並列化が容易でない事である。深層学習に代表される近年の有用なアルゴリズムの多くは多数のコアで効率良く動作する並列アルゴリズムである。探索アルゴリズムは有用であるが、並列化が難しいと思われている。
    この2点を解決し、機械学習+探索の枠組みを広く普及させることを目指す。

  • 機械学習と探索の協調による高性能最適化アルゴリズム

    Grant number:20H04251  2020 - 2024

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (B)

      More details

    Authorship:Principal investigator  Grant type:Scientific research funding

  • AI技術に基づく短期交通予測手法と総合的な交通需要マネジメントの研究開発

    2018 - 2020

    新道路学術会議

      More details

    Authorship:Coinvestigator(s)  Grant type:Contract research

  • 複合的・階層的な自動チューニングを実現する数理基盤手法の研究とライブラリの開発

    2015.4 - 2018.3

      More details

    Authorship:Principal investigator 

    自動チューニングは、ソフトウェアが内包するパラメタを自ら調整し、多様な条件下で良好な性能を達成することを目指す技術である。従来、複数のパラメタの調整が必要な場合、網羅試行か経験的枝刈りが広く用いられてきたが、本研究では、ベイズ統計に基づき、現実的に有効かつ漸近的に最適解を導く数理的手法を目指した。
    従来研究の調査により、線形モデルと相関モデルが使われており、両者は同時に使うこともできることを示した。そのようなモデルを記述から、事前情報と実測結果から性能モデルを構築するプログラムを自動生成する仕組みを構築した。また、自動チューニング数理ライブラリの多様な計算に適用し、その有効性を確認した。

  • 複合的・階層的な自動チューニングを実現する数理基盤手法の研究とライブラリの開発

    Grant number:15H02708  2015 - 2017

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (B)

      More details

    Authorship:Coinvestigator(s)  Grant type:Scientific research funding

  • 連続・離散ハイブリッド領域のための区間制約プログラミング技術

    Grant number:15K15968  2015 - 2017

    Grants-in-Aid for Scientific Research  Grant-in-Aid for Young Scientists (B)

      More details

    Grant type:Scientific research funding

  • 1万コア以上を用いる並列探索アルゴリズムで囲碁名人に勝つ

    Grant number:25700038  2013 - 2016

    Grants-in-Aid for Scientific Research  Grant-in-Aid for Young Scientists (A)

      More details

    Authorship:Principal investigator  Grant type:Scientific research funding

  • 汎用自動チューニング機構を実現するためのソフトウェア基盤の研究

    Grant number:23240005  2011 - 2014

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (A)

      More details

    Grant type:Scientific research funding

  • モンテカルロ木探索の汎用的高速化手法の研究

    Grant number:23700158  2011 - 2012

    Grants-in-Aid for Scientific Research  Grant-in-Aid for Young Scientists (B)

      More details

    Authorship:Principal investigator  Grant type:Scientific research funding

  • 生体及び人工物の高精度・高信頼認識技術の研究

    Grant number:19200006  2007 - 2009

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (A)

      More details

    Authorship:Coinvestigator(s)  Grant type:Scientific research funding

▼display all

Educational Activities

  • 情報基盤研究開発センターが本務だが、システム情報科学府を兼務し、大学院講義に加えて理学部物理学科情報理学コース、工学部電気情報工学科の講義を一部担当している。

Class subject

  • グラフ探索アルゴリズムⅡ

    2024.6 - 2024.8   Summer quarter

  • Graph Search Algorithms II

    2024.6 - 2024.8   Summer quarter

  • 【通年】情報理工学講究

    2024.4 - 2025.3   Full year

  • 【通年】情報理工学演習

    2024.4 - 2025.3   Full year

  • 【通年】情報理工学研究Ⅰ

    2024.4 - 2025.3   Full year

  • 情報理工学論議Ⅰ

    2024.4 - 2024.9   First semester

  • 情報理工学論述Ⅰ

    2024.4 - 2024.9   First semester

  • 情報理工学読解

    2024.4 - 2024.9   First semester

  • Graph Search Algorithms I

    2024.4 - 2024.6   Spring quarter

  • グラフ探索アルゴリズムⅠ

    2024.4 - 2024.6   Spring quarter

  • 並列アルゴリズムⅡ

    2023.12 - 2024.2   Winter quarter

  • 並列アルゴリズム

    2023.10 - 2024.3   Second semester

  • 並列アルゴリズムⅠ

    2023.10 - 2023.12   Fall quarter

  • グラフ探索アルゴリズムⅡ

    2023.6 - 2023.8   Summer quarter

  • Graph Search Algorithms II

    2023.6 - 2023.8   Summer quarter

  • 【通年】情報理工学研究Ⅰ

    2023.4 - 2024.3   Full year

  • 【通年】情報理工学講究

    2023.4 - 2024.3   Full year

  • 【通年】情報理工学演習

    2023.4 - 2024.3   Full year

  • 情報理工学読解

    2023.4 - 2023.9   First semester

  • 情報理工学論議Ⅰ

    2023.4 - 2023.9   First semester

  • 情報理工学論述Ⅰ

    2023.4 - 2023.9   First semester

  • Graph Search Algorithms I

    2023.4 - 2023.6   Spring quarter

  • グラフ探索アルゴリズムⅠ

    2023.4 - 2023.6   Spring quarter

  • サイバーセキュリティ基礎論

    2023.4 - 2023.6   Spring quarter

  • サイバーセキュリティ基礎論

    2023.4 - 2023.6   Spring quarter

  • 情報科学講究

    2022.10 - 2023.3   Second semester

  • 情報理工学論議Ⅱ

    2022.10 - 2023.3   Second semester

  • 情報理工学論述Ⅱ

    2022.10 - 2023.3   Second semester

  • 情報理工学演示

    2022.10 - 2023.3   Second semester

  • プログラミング技法

    2022.10 - 2023.3   Second semester

  • 並列アルゴリズム

    2022.10 - 2023.3   Second semester

  • 並列アルゴリズム

    2022.10 - 2023.3   Second semester

  • 情報科学講究

    2022.10 - 2023.3   Second semester

  • グラフ探索アルゴリズムⅡ

    2022.6 - 2022.8   Summer quarter

  • 情報理工学研究Ⅰ

    2022.4 - 2023.3   Full year

  • 情報理工学講究

    2022.4 - 2023.3   Full year

  • 情報理工学演習

    2022.4 - 2023.3   Full year

  • 情報理工学読解

    2022.4 - 2022.9   First semester

  • 情報理工学論議Ⅰ

    2022.4 - 2022.9   First semester

  • 情報理工学論述Ⅰ

    2022.4 - 2022.9   First semester

  • サイバーセキュリティ基礎論

    2022.4 - 2022.6   Spring quarter

  • グラフ探索アルゴリズムⅠ

    2022.4 - 2022.6   Spring quarter

  • サイバーセキュリティ基礎論

    2022.4 - 2022.6   Spring quarter

  • 情報科学講究

    2021.10 - 2022.3   Second semester

  • 国際科学特論Ⅱ

    2021.10 - 2021.12   Fall quarter

  • 情報科学講究

    2021.10 - 2021.12   Fall quarter

  • 国際科学特論Ⅱ

    2021.10 - 2021.12   Fall quarter

▼display all

FD Participation

  • 2023.10   Role:Participation   Title:【シス情FD】価値創造型半導体人材育成センターについて

    Organizer:[Undergraduate school/graduate school/graduate faculty]

  • 2023.9   Role:Participation   Title:【シス情FD】Top10%論文/Top10%ジャーナルとは何か: 傾向と対策

    Organizer:[Undergraduate school/graduate school/graduate faculty]

  • 2023.4   Role:Participation   Title:【シス情FD】若手教員による研究紹介⑧

    Organizer:[Undergraduate school/graduate school/graduate faculty]

  • 2023.3   Role:Participation   Title:【シス情FD】独・蘭・台湾での産学連携を垣間見る-Industy 4.0・量子コンピューティング・先端半導体-

    Organizer:[Undergraduate school/graduate school/graduate faculty]

  • 2023.1   Role:Participation   Title:【シス情FD】若手教員による研究紹介⑦

    Organizer:[Undergraduate school/graduate school/graduate faculty]

  • 2022.6   Role:Participation   Title:【シス情FD】電子ジャーナル等の今後について

    Organizer:[Undergraduate school/graduate school/graduate faculty]

  • 2022.5   Role:Participation   Title:【シス情FD】若手教員による研究紹介④「量子コンピュータ・システム・アーキテクチャの研究~道具になることを目指して~」

    Organizer:[Undergraduate school/graduate school/graduate faculty]

  • 2022.4   Role:Participation   Title:【シス情FD】第4期中期目標・中期計画等について

    Organizer:[Undergraduate school/graduate school/graduate faculty]

  • 2022.4   Role:Participation   Title:令和4年度 第1回全学FD(新任教員の研修)The 1st All-University FD (training for new faculty members) in FY2022

    Organizer:University-wide

  • 2022.3   Role:Participation   Title:全学FD:メンタルヘルス講演会

    Organizer:University-wide

  • 2022.3   Role:Participation   Title:新M2Bシステムの使い方 ~新機能を中心に紹介します~(3/17)

    Organizer:University-wide

  • 2022.1   Role:Participation   Title:【シス情FD】シス情関連の科学技術に対する国の政策動向(に関する私見)

    Organizer:[Undergraduate school/graduate school/graduate faculty]

  • 2021.10   Role:Participation   Title:【シス情FD】熊本高専と九大システム情報との交流・連携に向けて ー 3年半で感じた高専の実像 ー

    Organizer:[Undergraduate school/graduate school/graduate faculty]

  • 2021.9   Role:Participation   Title:M2B学習支援システム講習会★初級・中上級編★

    Organizer:University-wide

▼display all

Visiting, concurrent, or part-time lecturers at other universities, institutions, etc.

  • 2021  東京農工大学工学部  Classification:Part-time lecturer  Domestic/International Classification:Japan 

    Semester, Day Time or Duration:電子情報工学特別講義II (集中講義)

Social Activities

  • これで書ける!コンピュータ囲碁講習会 第0回 概論編

    電気通信大学  電気通信大学(東京都調布市)  2015.7

     More details

    Audience:General, Scientific, Company, Civic organization, Governmental agency

    Type:Lecture

  • CESA Developer's Conference (CEDEC2009) 講演 「モンテカルロ法によるゲームAIの可能性」

    社団法人コンピュータエンターテインメント協会(CESA)  パシフィコ横浜 会議センター(横浜西区みなとみらい)  2009.9

     More details

    Audience:General, Scientific, Company, Civic organization, Governmental agency

    Type:Lecture

  • 公開講座 囲碁AIにおける革命「モンテカルロ木探索」とは何か?

    DigraJapan(日本デジタルゲーム学会)  東京大学(東京都文京区)  2008.11

     More details

    Audience:General, Scientific, Company, Civic organization, Governmental agency

    Type:Lecture

Media Coverage

  • AIで有機化合物探索

    日刊工業新聞  2018.4

     More details

    AIで有機化合物探索

Travel Abroad

  • 2011.2 - 2011.5

    Staying countory name 1:Canada   Staying institution name 1:アルバータ大学