Spontaneous Evolution of Structural Modularity in Robot Neural Network Controllers

被引：0
|
作者：
Bongard, Josh C. [1 ]
机构：
[1] Univ Vermont, Dept Comp Sci, Burlington, VT 05405 USA
来源：
GECCO-2011: PROCEEDINGS OF THE 13TH ANNUAL GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE | 2011年
关键词：
Evolutionary Robotics; Modularity; Evolutionary Algorithms; BEHAVIOR;
D O I：
暂无
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
In order to evolve large robot controllers for increasingly complex tasks, fully connected neural networks are not feasible. However, manually designing sparse neural connectivity is not intuitive, and thus should be placed under evolutionary control. Here I show how spontaneous structural modularity can arise in the connectivity of evolved robot controllers if the controllers are boolean networks, and are selected to converge on point attractors that correspond to successful robot behaviors.
引用
收藏
页码：251 / 258
页数：8
相关论文
共 50 条
  • [1] Spontaneous evolution of modularity and network motifs
    Kashtan, N
    Alon, U
    PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2005, 102 (39) : 13773 - 13778
  • [2] Evolution of Neural Controllers for Robot Formation
    Capi, Genci
    6TH INTERNATIONAL CONFERENCE ON SOFT COMPUTING AND INTELLIGENT SYSTEMS, AND THE 13TH INTERNATIONAL SYMPOSIUM ON ADVANCED INTELLIGENT SYSTEMS, 2012, : 1275 - 1280
  • [3] Noise and the Evolution of Neural Network Modularity
    Hoverstad, Boye Annfelt
    ARTIFICIAL LIFE, 2011, 17 (01) : 33 - 50
  • [4] The evolution of neural network controllers for the targeted swing of cockroach-like robot
    Fife, A
    Laksanacharoen, S
    Quinn, RD
    Beer, RD
    Ritzmann, RE
    PROCEEDING OF THE 2002 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-3, 2002, : 929 - 933
  • [5] Artificial Neural Network Controllers for Biped Robot
    Rai, J. K.
    Singh, V. P.
    Tewari, R. P.
    Chandra, D.
    2012 2ND INTERNATIONAL CONFERENCE ON POWER, CONTROL AND EMBEDDED SYSTEMS (ICPCES 2012), 2012,
  • [6] Modularity and Sparsity: Evolution of Neural Net Controllers in Physically Embodied Robots
    Livingston, Nicholas
    Bernatskiy, Anton
    Livingston, Kenneth
    Smith, Marc L.
    Schwarz, Jodi
    Bongard, Joshua C.
    Wallach, David
    Long, John H., Jr.
    FRONTIERS IN ROBOTICS AND AI, 2016, 3
  • [7] Neural network - Based controllers for mobile robot
    de Sousa, C
    Hemerly, EM
    SIXTH BRAZILIAN SYMPOSIUM ON NEURAL NETWORKS, VOL 1, PROCEEDINGS, 2000, : 50 - 55
  • [8] Evolving Robot Morphology Facilitates the Evolution of Neural Modularity and Evolvability
    Bongard, Josh C.
    Bernatskiy, Anton
    Livingston, Ken
    Livingston, Nicholas
    Long, John
    Smith, Marc
    GECCO'15: PROCEEDINGS OF THE 2015 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, 2015, : 129 - 136
  • [9] Evolution of visually-guided approach behaviour in recurrent artificial neural network robot controllers
    Biro, Z
    Ziemke, T
    FROM ANIMALS TO ANIMATS 5, 1998, : 73 - 76
  • [10] On 'parts' and 'wholes' of adaptive behavior: Functional modularity and diachronic structure in recurrent neural robot controllers
    Ziemke, T
    FROM ANIMALS TO ANIMATS 6, 2000, : 115 - 124
12345