• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Development of evolutionary computation techniques to realize human-out-of-the-loop

Research Project

Project/Area Number 19H04179
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Review Section Basic Section 61040:Soft computing-related
Research InstitutionUniversity of Tsukuba

Principal Investigator

Akimoto Youhei  筑波大学, システム情報系, 准教授 (20709654)

Project Period (FY) 2019-04-01 – 2024-03-31
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥17,160,000 (Direct Cost: ¥13,200,000、Indirect Cost: ¥3,960,000)
Fiscal Year 2023: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
Fiscal Year 2022: ¥3,770,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥870,000)
Fiscal Year 2021: ¥3,380,000 (Direct Cost: ¥2,600,000、Indirect Cost: ¥780,000)
Fiscal Year 2020: ¥3,380,000 (Direct Cost: ¥2,600,000、Indirect Cost: ¥780,000)
Fiscal Year 2019: ¥3,900,000 (Direct Cost: ¥3,000,000、Indirect Cost: ¥900,000)
Keywords進化計算 / マルチフィデリティ最適化 / ワーストケース最適化 / 制約付き最適化 / 深層生成モデル / 収束解析 / Human-out-of-the-loop / 不確実性 / Min-Max最適化 / 多目的最適化 / 深層学習 / 生成モデル / 非凸制約 / シミュレーションベース最適化 / ナッシュ均衡解 / マルチフィデリティ / アルゴリズム選択 / シミュレーションベース最適化
Outline of Research at the Start

ブラックボックス最適化技術は,これまで技術者のノウハウや試行錯誤に基づいていたパラメータ調整を人手を介さずに実現する基盤技術であるが,最適化結果に直結する重要な意思決定は依然として問題設計者のノウハウと試行錯誤を要する.本研究では,最適化プロセス全体の自動化を実現するために,進化計算と呼ばれる汎用的な最適化技術を応用した新しい最適化の枠組みを提案する.これにより,問題設計者が被っていた試行錯誤に基づく意思決定の負担が大幅に削減され,最適化プロセス全体の抜本的な効率化が期待される.

Outline of Final Research Achievements

To examine the question, “Can the optimization process be automated by delegating decision-making to optimization methods?”, we developed a framework for automated decision-making for items that problem designers face in dealing with simulation-based optimization and that are directly related to the optimization results. Specifically, we proposed an automatic simulation accuracy selection mechanism based on rank correlation coefficients, proposed an optimization method that guarantees worst-case performance, analyzed the behavior of evolution strategies using a surrogate function, considered termination conditions by analyzing the convergence rate of evolution strategies, automatically constructed design variables using deep generative models for constrained optimization, and developed an efficient and effective optimization method.

Academic Significance and Societal Importance of the Research Achievements

与えられた解に対する評価のみを返す評価器を通して最適化を実現するブラックボックス最適化技術は,これまで技術者のノウハウや試行錯誤に基づいていたパラメータ調整を人手を介さずに実現する基盤技術である.しかし,問題の定式化や与えられた解の目的関数値や制約違反量を計算する評価器の設計は問題設計者(人)の役割であり,最適化法の選択やパラメータの調整は最適化実施者(人)の役割であるため,システム全体の自動化を実現には至っていなかった.本研究の成果により,これまで最適化を繰り返しながら試行錯誤的に行われていた上述の意思決定への人の介入が不要となり,問題設計者の負担が大幅に削減されると期待される.

Report

(6 results)
  • 2023 Annual Research Report   Final Research Report ( PDF )
  • 2022 Annual Research Report
  • 2021 Annual Research Report
  • 2020 Annual Research Report
  • 2019 Annual Research Report
  • Research Products

    (37 results)

All 2024 2023 2022 2021 2020 2019 Other

All Int'l Joint Research (4 results) Journal Article (24 results) (of which Int'l Joint Research: 3 results,  Peer Reviewed: 24 results,  Open Access: 17 results) Presentation (8 results) Remarks (1 results)

  • [Int'l Joint Research] Inria(フランス)

    • Related Report
      2021 Annual Research Report
  • [Int'l Joint Research] Inria(フランス)

    • Related Report
      2020 Annual Research Report
  • [Int'l Joint Research] Inria(フランス)

    • Related Report
      2019 Annual Research Report
  • [Int'l Joint Research] RUB(ドイツ)

    • Related Report
      2019 Annual Research Report
  • [Journal Article] Convergence Rate of the (1+1)-ES on Locally Strongly Convex and Lipschitz Smooth Functions2024

    • Author(s)
      Morinaga Daiki、Fukuchi Kazuto、Sakuma Jun、Akimoto Youhei
    • Journal Title

      IEEE Transactions on Evolutionary Computation

      Volume: 28 Issue: 2 Pages: 501-515

    • DOI

      10.1109/tevc.2023.3266955

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Covariance Matrix Adaptation Evolutionary Strategy with Worst-Case Ranking Approximation for Min-Max Optimization and Its Application to Berthing Control Tasks2023

    • Author(s)
      Miyagi Atsuhiro、Miyauchi Yoshiki、Maki Atsuo、Fukuchi Kazuto、Sakuma Jun、Akimoto Youhei
    • Journal Title

      ACM Transactions on Evolutionary Learning and Optimization

      Volume: 3 Issue: 2 Pages: 1-32

    • DOI

      10.1145/3603716

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Trade-off Between Robustness and Worst-Case Performance in Min-Max Optimization2023

    • Author(s)
      Edo Hinata、Miyauchi Yoshiki、Maki Atsuo、Akimoto Youhei
    • Journal Title

      GECCO '23: Proceedings of the Genetic and Evolutionary Computation Conference

      Volume: - Pages: 1339-1347

    • DOI

      10.1145/3583131.3590362

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed
  • [Journal Article] CMA-ES with Learning Rate Adaptation: Can CMA-ES with Default Population Size Solve Multimodal and Noisy Problems?2023

    • Author(s)
      Nomura Masahiro、Akimoto Youhei、Ono Isao
    • Journal Title

      GECCO '23: Proceedings of the Genetic and Evolutionary Computation Conference

      Volume: - Pages: 839-847

    • DOI

      10.1145/3583131.3590358

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Adaptive scenario subset selection for worst-case optimization and its application to well placement optimization2023

    • Author(s)
      Miyagi Atsuhiro、Fukuchi Kazuto、Sakuma Jun、Akimoto Youhei
    • Journal Title

      Applied Soft Computing

      Volume: 133 Pages: 109842-109842

    • DOI

      10.1016/j.asoc.2022.109842

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Analysis of Surrogate-Assisted Information-Geometric Optimization Algorithms2022

    • Author(s)
      Akimoto Youhei
    • Journal Title

      Algorithmica

      Volume: - Issue: 1 Pages: 33-63

    • DOI

      10.1007/s00453-022-01087-8

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Adaptive Ranking-based Constraint Handling for Explicitly Constrained Black-Box Optimization2022

    • Author(s)
      Sakamoto Naoki、Akimoto Youhei
    • Journal Title

      Evolutionary Computation

      Volume: 30-2 Issue: 4 Pages: 503-529

    • DOI

      10.1162/evco_a_00310

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Explicitly Constrained Black-Box Optimization With Disconnected Feasible Domains Using Deep Generative Models2022

    • Author(s)
      Sakamoto Naoki、Sato Rei、Fukuchi Kazuto、Sakuma Jun、Akimoto Youhei
    • Journal Title

      IEEE Access

      Volume: 10 Pages: 117501-117514

    • DOI

      10.1109/access.2022.3219979

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Global Linear Convergence of Evolution Strategies on More than Smooth Strongly Convex Functions2022

    • Author(s)
      Akimoto Youhei、Auger Anne、Glasmachers Tobias、Morinaga Daiki
    • Journal Title

      SIAM Journal on Optimization

      Volume: 32 Issue: 2 Pages: 1402-1429

    • DOI

      10.1137/20m1373815

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Few-Shot Image-to-Semantics Translation for Policy Transfer in Reinforcement Learning2022

    • Author(s)
      Sato Rei、Fukuchi Kazuto、Sakuma Jun、Akimoto Youhei
    • Journal Title

      Proceedings of the International Joint Conference on Neural Networks

      Volume: - Pages: 01-10

    • DOI

      10.1109/ijcnn55064.2022.9892464

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Black-box min-max continuous optimization using CMA-ES with worst-case ranking approximation2022

    • Author(s)
      Miyagi Atsuhiro、Fukuchi Kazuto、Sakuma Jun、Akimoto Youhei
    • Journal Title

      Proceedings of the Genetic and Evolutionary Computation Conference

      Volume: - Pages: 823-831

    • DOI

      10.1145/3512290.3528702

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Monotone improvement of information-geometric optimization algorithms with a surrogate function2022

    • Author(s)
      Akimoto Youhei
    • Journal Title

      Proceedings of the Genetic and Evolutionary Computation Conference

      Volume: - Pages: 1354-1362

    • DOI

      10.1145/3512290.3528690

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] An ODE method to prove the geometric convergence of adaptive stochastic algorithms2022

    • Author(s)
      Akimoto Youhei、Auger Anne、Hansen Nikolaus
    • Journal Title

      Stochastic Processes and their Applications

      Volume: 145 Pages: 269-307

    • DOI

      10.1016/j.spa.2021.12.005

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Convergence rate of the (1+1)-evolution strategy with success-based step-size adaptation on convex quadratic functions2021

    • Author(s)
      Morinaga Daiki、Fukuchi Kazuto、Sakuma Jun、Akimoto Youhei
    • Journal Title

      Proceedings of the Genetic and Evolutionary Computation Conference (GECCO '21)

      Volume: - Pages: 1169-1177

    • DOI

      10.1145/3449639.3459289

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Level generation for angry birds with sequential VAE and latent variable evolution2021

    • Author(s)
      Tanabe Takumi、Fukuchi Kazuto、Sakuma Jun、Akimoto Youhei
    • Journal Title

      Proceedings of the Genetic and Evolutionary Computation Conference (GECCO '21)

      Volume: - Pages: 1052-1060

    • DOI

      10.1145/3449639.3459290

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Saddle point optimization with approximate minimization oracle2021

    • Author(s)
      Akimoto Youhei
    • Journal Title

      Proceedings of the Genetic and Evolutionary Computation Conference (GECCO '21)

      Volume: - Pages: 493-501

    • DOI

      10.1145/3449639.3459266

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Adaptive scenario subset selection for min-max black-box continuous optimization2021

    • Author(s)
      Miyagi Atsuhiro、Fukuchi Kazuto、Sakuma Jun、Akimoto Youhei
    • Journal Title

      Proceedings of the Genetic and Evolutionary Computation Conference (GECCO '21)

      Volume: - Pages: 697-705

    • DOI

      10.1145/3449639.3459291

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Diagonal Acceleration for Covariance Matrix Adaptation Evolution Strategies2020

    • Author(s)
      Akimoto Y.、Hansen N.
    • Journal Title

      Evolutionary Computation

      Volume: 28 Issue: 3 Pages: 405-435

    • DOI

      10.1162/evco_a_00260

    • NAID

      120007168383

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Deep generative model for non-convex constraint handling2020

    • Author(s)
      Sakamoto Naoki、Semmatsu Eiji、Fukuchi Kazuto、Sakuma Jun、Akimoto Youhei
    • Journal Title

      Proceedings of the 2020 Genetic and Evolutionary Computation Conference (GECCO '20)

      Volume: - Pages: 636-644

    • DOI

      10.1145/3377930.3390170

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Multi-fidelity Optimization Approach Under Prior and Posterior Constraints and Its Application to Compliance Minimization2020

    • Author(s)
      Akimoto Youhei、Sakamoto Naoki、Ohtani Makoto
    • Journal Title

      Parallel Problem Solving from Nature - PPSN XVI

      Volume: - Pages: 81-94

    • DOI

      10.1007/978-3-030-58112-1_6

    • ISBN
      9783030581114, 9783030581121
    • Related Report
      2020 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Adaptive objective selection for multi-fidelity optimization2019

    • Author(s)
      Akimoto Youhei、Shimizu Takuma、Yamaguchi Takahiro
    • Journal Title

      Proceedings of the Genetic and Evolutionary Computation Conference

      Volume: - Pages: 880-888

    • DOI

      10.1145/3321707.3321709

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Adaptive ranking based constraint handling for explicitly constrained black-box optimization2019

    • Author(s)
      Sakamoto Naoki、Akimoto Youhei
    • Journal Title

      Proceedings of the Genetic and Evolutionary Computation Conference

      Volume: - Pages: 700-708

    • DOI

      10.1145/3321707.3321717

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Well placement optimization under geological statistical uncertainty2019

    • Author(s)
      Miyagi Atsuhiro、Akimoto Youhei、Yamamoto Hajime
    • Journal Title

      Proceedings of the Genetic and Evolutionary Computation Conference

      Volume: - Pages: 1284-1292

    • DOI

      10.1145/3321707.3321736

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Generalized drift analysis in continuous domain2019

    • Author(s)
      Morinaga Daiki、Akimoto Youhei
    • Journal Title

      Proceedings of the 15th ACM/SIGEVO Conference on Foundations of Genetic Algorithms

      Volume: - Pages: 13-24

    • DOI

      10.1145/3299904.3340303

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Presentation] 障害物を考慮した船舶着桟制御方策のDomain Randomizationによる獲得の試み2023

    • Author(s)
      河上 幸樹
    • Organizer
      2023年度人工知能学会全国大会(第37回)
    • Related Report
      2023 Annual Research Report
  • [Presentation] 解の評価が高コストな場合における多様性最適化法の提案と評価2022

    • Author(s)
      三村遼,秋本洋平
    • Organizer
      進化計算学会研究会
    • Related Report
      2021 Annual Research Report
  • [Presentation] モデル化誤差に対してロバストな着桟制御に向けた検討2021

    • Author(s)
      秋本 洋平
    • Organizer
      日本船舶海洋工学会講演会
    • Related Report
      2021 Annual Research Report
  • [Presentation] Sequential Variational Autoencoderを用いたAngry Birdsのステージ生成2021

    • Author(s)
      田邊 拓実 , 福地 一斗 , 佐久間 淳 , 秋本 洋平
    • Organizer
      人工知能学会全国大会
    • Related Report
      2021 Annual Research Report
  • [Presentation] 深層生成モデルによる弱パレート解集合の近似2021

    • Author(s)
      江戸 陽向 , 濱田 直希 , 福地 一斗 , 佐久間 淳 , 秋本 洋平
    • Organizer
      人工知能学会全国大会
    • Related Report
      2021 Annual Research Report
  • [Presentation] 複数の損失関数を用いた深層生成モデルの訓練と制約付きブラックボックス最適化への適用2021

    • Author(s)
      阪本 直気 , 佐藤 怜 , 福地 一斗 , 佐久間 淳 , 秋本 洋平
    • Organizer
      人工知能学会全国大会
    • Related Report
      2021 Annual Research Report
  • [Presentation] (1+1)-Evolution Strategyの凸二次関数における収束速度の導出2021

    • Author(s)
      森永 大貴 , 福地 一斗 , 佐久間 淳 , 秋本 洋平
    • Organizer
      人工知能学会全国大会
    • Related Report
      2021 Annual Research Report
  • [Presentation] 非凸制約付きシミュレーションベース最適化における深層生成モデルの新たな活用法2019

    • Author(s)
      阪本 直気, 佐久間 淳,秋本 洋平
    • Organizer
      計測自動制御学会 システム・情報部門 学術講演会 2019
    • Related Report
      2019 Annual Research Report
  • [Remarks] Black-Box Optimization Lab.

    • URL

      https://www.bbo.cs.tsukuba.ac.jp/

    • Related Report
      2021 Annual Research Report 2020 Annual Research Report

URL: 

Published: 2019-04-18   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi