"CAM-BRAIN"
ATR's ARTIFICIAL BRAIN PROJECT
An Overview

Hugo de Garis (1), Michael Korkin (2), Felix Gers (3), Norberto Eiji Nawa (1), Michael Hough (4)
(1) Brain Builder Group,
Evolutionary Systems Department,
ATR Human Information Processing Research Labs.
2-2 Hikari-dai, Seika-cho, Soraku-gun, Kyoto-fu, 619-0288, Japan
Tel: +81-774-95-1079, Fax: +81-774-95-1008
degaris@hip.atr.co.jp, xnawa@hip.atr.co.jp,
http://www.hip.atr.co.jp/degaris, http://www.hip.atr.co.jp/xnawa
(2) Genobyte Inc.,
1503 Spruce Street, Suite 3, Boulder, Colorado 80302, USA
tel. +1-303-545-6790, fax. +1-303-415-1109
korkin@genobyte.com, http://www.genobyte.com
(3) Istituto Dalle Molle di Studi sull'Intelligenza Artificiale (IDSIA),
Corso Elvezia 36, CH-6900, Lugano, Switzerland
tel. +41-91-911-9838, fax. +41-91-911-9839
felix@idsia.ch, http://www.idsia.ch/felix/
(4) Computer Science Department
Stanford University, Stanford, 94305, CA, USA
mhough@cs.stanford.edu, http://www.stanford.edu/mhough

Abstract:

This paper provides an overview of ATR's Artificial Brain (CAM-Brain) Project, which aims to build an artificial brain containing some billion artificial neurons by the year 2001, and to use this brain to control the behaviors of a robot kitten "Robokoneko".

 
• Introduction
• The CoDi-1Bit Neural Network Model
• The CAM-Brain Machine (CBM)
  • CBM Overview
  • CBM Architecture
    • Cellular Automata Module
    • Genotype and Phenotype Memory
    • Fitness Evaluation Unit
    • Genetic Algorithm Unit
    • Module Interconnection Memory
    • External Interface
• The Spike Interval Information Coding Representation, "SIIC"
  • Choosing a Representation for the CoDi-1Bit Signaling
  • The SIIC Convolution Algorithm
    • Simplified Example
• The "Hough Spiker Algorithm" (HSA) for Deconvolution
• "Robokoneko", the Kitten Robot
• A Sampler of CoDi-1Bit Evolved Neural Net Modules
  • XOR Module
  • Timer Module
  • Multiple Timer Module
  • Switchable Dual Function Module
  • Pattern Detector Module
  • Hubel-Wiesel Line Motion Detector Module
• Ideas for Interesting Future CoDi Modules to be Evolved
  • The "Dagwood Approach" for a Multi-Test Module
  • Modules Which Learn
  • From Multi Module Systems to Artificial Brains
• Conclusions
• References
• About this document ...