Amoeba-based neurocomputing with chaotic dynamics

Masashi Aono, Masahiko Hara, Kazuyuki Aihara

Research output: Contribution to journalReview article

42 Citations (Scopus)

Abstract

An amoeba that freely change its planar shape inside the stellar carrier structure on an agar plate was employed for neuralcomputing with chaotic dynamics. A neural network model with state transition represented by the amoeba's photoaviodance-based shape deformation under optical feedback control was also implemented. Optical feedback automatically is found to update illumination according to recurrent neural network dynamics while each neuron is activated or inactivated when the weighted sum of inputs from other neurons exceed a certain threshold. The symmetry-broken oscillation patterns of amoeba imply that the system operates as a logical circuit with computational university by properly altering the number of neurons, thresholds, and weights. Collectively interacting amoeba proteins are found to generate chaotic dynamics capable of amplifying intrinsic microscopic fluctuations to destabilize neural macroscopic conditions.

Original languageEnglish
Pages (from-to)69-72
Number of pages4
JournalCommunications of the ACM
Volume50
Issue number9
DOIs
Publication statusPublished - 2007 Sep 1
Externally publishedYes

    Fingerprint

ASJC Scopus subject areas

  • Computer Science(all)

Cite this