Cristin-resultat-ID: 1979266
Sist endret: 31. mars 2022, 13:44
Resultat
Rapport
2021

Towards Self-organized Control: Using Neural Cellular Automata to Robustly Control a Cart-pole Agent

Bidragsytere:
  • Alexandre Variengien
  • Sidney Pontes Filho
  • Tom Eivind Glover og
  • Stefano Nichele

Utgiver/serie

Utgiver

Innovations in Machine Intelligence, Crosslabs, Vol. 1/2021

Om resultatet

Rapport
Publiseringsår: 2021
Volum: 1
Antall sider: 14

Klassifisering

Fagfelt (NPI)

Fagfelt: IKT
- Fagområde: Realfag og teknologi

Beskrivelse Beskrivelse

Tittel

Towards Self-organized Control: Using Neural Cellular Automata to Robustly Control a Cart-pole Agent

Sammendrag

Neural cellular automata (Neural CA) are a recent framework used to model biological phenomena emerging from multicellular organisms. In these systems, artificial neural networks are used as update rules for cellular automata. Neural CA are end-to-end differentiable systems where the parameters of the neural network can be learned to achieve a particular task. In this work, we used neural CA to control a cart-pole agent. The observations of the environment are transmitted in input cells while the values of output cells are used as a readout of the system. We trained the model using deep-Q learning where the states of the output cells were used as the Q-value estimates to be optimized. We found that the computing abilities of the cellular automata were maintained over several hundreds of thousands of iterations, producing an emergent stable behavior in the environment it controls for thousands of steps. Moreover, the system demonstrated life-like phenomena such as a developmental phase, regeneration after damage, stability despite a noisy environment, and robustness to unseen disruption such as input deletion.

Bidragsytere

Alexandre Variengien

  • Tilknyttet:
    Forfatter

Sidney Pontes Filho

  • Tilknyttet:
    Forfatter
    ved Institutt for datateknologi og informatikk ved Norges teknisk-naturvitenskapelige universitet

Tom Glover

Bidragsyterens navn vises på dette resultatet som Tom Eivind Glover
  • Tilknyttet:
    Forfatter
    ved Institutt for informasjonsteknologi ved OsloMet - storbyuniversitetet

Stefano Nichele

  • Tilknyttet:
    Forfatter
    ved Institutt for informasjonsteknologi ved OsloMet - storbyuniversitetet
1 - 4 av 4