ATR’s Artificial Brain Project: CAM-Brain Machine (CBM) and Robot Kitten (Robokoneko) Issues
This paper presents some ongoing issues concerning ATR’s Artificial Brain (CAM-Brain) Project. The CAM-Brain Project evolves 3D cellular automata (CA) based neural networks directly in FPGA electronics at electronic speeds in special hardware called a CAM-Brain Machine (CBM). The CBM updates the CA cells at a rate of 150 Billion a second, and can perform a full run of a genetic algorithm (GA) in about 1 second. 32K of these evolved circuits (modules) are then assembled (in a large HAM space updated in real-time by the CBM) into humanly defined architectures to make an artificial brain to control a robot kitten (“Robokoneko”). The paper presents and discusses the latest design decisions for the CBM and the kitten robot, and maps out future plans aimed at having an artificial-brain-controlled robot kitten playing in the ATR, labs by early 2001. A world wide, many membered, internet-videophone and electronic pen based, brain-architectural design and brainstorming group will be essential for distributing the design and evolution of the 32K modules. Designing and building an artificial brain within the next three years will be a major conceptual and managerial challenge.
KeywordsCellular Automaton Spike Train Cellular Automaton Neural Module Blank Cell
Unable to display preview. Download preview PDF.
- DE Garis, H. An artificial brain: ATR’s cam-brain project aims to build/evolve an artificial brain with a million neural net modules inside a trillion cell cellular automata machine. New Generation Computing Journal 12 2 (July 1994).Google Scholar
- Gers, F., DE Garis, H., AND Korkin, M. Codi-1 Bit: A simplified cellular automata based neuron model. In Proceedings of AE97, Artificial Evolution Conference (October 1997).Google Scholar
- Korkin, M., DE Garis, H., Gers, F., AND Hemmi, H. CBM (CAM-Brain Machine): A hardware tool which evolves a neural net module in a fraction of a second and runs a million neuron artificial brain in real time. In Genetic Programming 1997: Proceedings of the Second Annual Conference (July 1997), J. R. Koza, K. Deb, M. Dorigo, D. B. Fogel, M. Garzon, H. Iba, and R. L. Riolo, Eds.Google Scholar
- Korkin, M., Nawa, N. E., AND DE Garis, H. A ’spike interval information coding’ representation for ATR’s CAM-brain machine (CBM). In Proceedings of the Second International Conference on Evolvable Systems: From Biology to Hardware (ICES′98) (September 1998), Springer-Verlag.Google Scholar
- Rieke, F., Warland, D., DE Ruyter VAN Steveninck, R., AND Bialek, W. Spikes: exploring the neural code. MIT Press/Bradford Books, Cambridge, MA, 1997.Google Scholar