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

2023 Fiscal Year Final Research Report

Research on motif structures for functional emergence in deep neuroevolution

Research Project

  • PDF
Project/Area Number 20H04253
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 InstitutionThe University of Tokyo

Principal Investigator

Iba Hitoshi  東京大学, 大学院情報理工学系研究科, 教授 (40302773)

Co-Investigator(Kenkyū-buntansha) 長谷川 禎彦  東京大学, 大学院情報理工学系研究科, 准教授 (20512354)
Project Period (FY) 2020-04-01 – 2023-03-31
Keywords進化計算 / 遺伝的アルゴリズム / 遺伝的プログラミング / 人工生命 / 群知能
Outline of Final Research Achievements

In this study, we aimed at function emergence based on motif structures in deep neuroevolution. Motif structures here are not simply similarities in the biological term DNA, but rather structures that originate from a common ancestor and cause the same functions, implying deeper information-theoretic features. In this study, we attempted to analyze the temporal developmental process of deep neuroevolution using nonlinear dynamical systems and information statistical mechanics methods. Based on the results, we controlled the temporal evolution of network motifs to achieve precise network expression and function emergence. The effectiveness of the proposed method is verified in a wide range of fields such as robotics, creative support and engineering optimization.

Free Research Field

進化計算

Academic Significance and Societal Importance of the Research Achievements

実世界応用として、ロボティクスやX線データによる危険物検出や医療用画像の解析を試みた.例えば医療応用では、X線動画からFBP法による再構築をした.医師の評価を踏まえ,X線動画からCT画像を生成する手法として有用であり得ることが確認された.具体的には、大学病院での定量的な評価が研究成果につながった.またロボティクス応用では、ソフトロボットに有用な構造と制御を同時に最適化する手法であるco-designというフレームワークを構築した.

URL: 

Published: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi