A "Spike Interval Information Coding" Representation for ATR's CAM-Brain Machine (CBM)

This paper reports on ongoing attempts to nd an eecient and eeective representation for the binary signaling of ATR's CAM-Brain Machine (CBM), using the so-called "CoDi-1Bit" model. The CBM is an Field Programmable Gate Array (FPGA) based hardware accelerator which updates 3D cellular automata (CA) cells at the rate of 100 billion a second, allowing a complete run of a genetic algorithm with tens of thousands of CA based neural net circuit growths and hardware compiled tness evaluations, all in about 1 second. It is hoped that using such a device, it will become practical to evolve 10,000s of neural net modules and then to assemble them into humanly deened RAM based artiicial brain architectures which can be run by the CBM in real time to control robots, e.g. a robot kitten. Before large numbers of modules can be assembled together, it is essential that the individual modules have a good functionality and evolvability. The "CoDi-1Bit" CA based neural network model uses 1 bit binary signaling, so a representation needs to be chosen based on this fact. This paper introduces and discusses the merits and demerits of a representation that we call "Spike Interval Information Coding" (SIIC). Simulation results using the SIIC representation method to evolve time dependent waveforms and simple functional modules are presented. The results indicate the suitability of the SIIC representation method to decode the bit streams generated by the CA based neural networks.