Holographic Automata for Ambient Immersive A. I. via Reservoir Computing
This work offers a novel mathematical framework for cellular automata that could enhance reservoir computing and enable new distributed AI systems, though it appears incremental in its core technical contribution.
The authors proved that cellular automata can be represented semilinearly using multiple convolution kernels, demonstrating this with edge-of-chaos rules like Turing-universal rule 110 for reservoir computing applications. They also proposed a 'Hologrid' system leveraging frequency-domain representations and existing wireless technologies for distributed computing.
We prove the existence of a semilinear representation of Cellular Automata (CA) with the introduction of multiple convolution kernels. Examples of the technique are presented for rules akin to the "edge-of-chaos" including the Turing universal rule 110 for further utilization in the area of reservoir computing. We also examine the significance of their dual representation on a frequency or wavelength domain as a superposition of plane waves for distributed computing applications including a new proposal for a "Hologrid" that could be realized with present Wi-Fi,Li-Fi technologies.