Robot Brain

Development versions of AjoChhand neural network simulator, based on our own Fractal Information theory where nested time cycle is the unit of information, self-assembly of reality spheres executes the programming and frequency wheel or the resonance chain creates the morphing of a physical system to execute a decision making. Its applications would be to verify the power of new kind of decision making that we have proposed. Note that the kind of frequency fractal computing that we have proposed cannot be implemented using conventional Turing machines. That means, no CMOS based circuits would be able to harness its powers fully. However, we always felt that even if we would fail completely to realize organic brain jelly an analogue for the brain’s neural network using

AjoChhand 1.0: Basic: A computer simulator based cortical column assembly that does everything a brain does, acquire sensory data as input, build the neural network, modify circuit geometry as required for learning and then from the modified circuit geometry, anyone can retrieve images and everything encoded into it. For example an input movie or song or story is saved into a set of frequencies and from a set of frequencies we retrieve entire images, songs, just like a “real” human brain.

AjoChhand 2.0: Management: We add multiple sensory modules side by side in a singular platform. The simulator incorporates an analogue of hippocampus for managing “language learning”, “creative painting”, “composing music”, “identifying speech” etc. Hippocampus based management and automated creation of necessary functional modules for information processing would be the key for this version. Add nucleus, and several local micro-management module options.

AjoChhand 3.0: Perception: The simulator incorporates the analogue of forebrain for generating collective perception among various learning networks. Higher-level perception is encoded even in a single cortical column assembly, however, not for the multiple functional modules. Here, higher-level perceptions are allowed to grow by itself.

AjoChhand 4.0: Nested Clock management: Including features to run time clocks for managing various tasks, thoughts, in the mid brain a special management center is kept to control millions of time bound jobs or thoughts. This is important for the simulator brain to do any work. Every single thoughts of entire brain is connected so that millions of tasks are executed with its virtual clock management

AjoChhand 5.0: Machine control: Introduce an artificial analogue of cerebellum, production house of millions of instructions for the sensors and machines connected to it for acquiring better data for analysis.

AjoChhand 6.0: Self-evolution of hardware: Introduce controls to add new sensory systems, global re-wiring management for correcting major circuit errors during self-learning. Introduce the control for self-de-fragmentation and arranging it, the brain would work 24X7, to improve its circuits and reorganize it by itself.

AjoChhand 7.0: Self-learning: Key rules for automated learning to be incorporated in the system, so that the simulator can automatically set control parameters for the automated data acquisition, does not become a random learner. Basic rules in the artificial analogue of the mid brain would enable the system to control all sensory and processing system.

AjoChhand 8.0: Simulate futuristic outcome: As a general concourse to information processing, the network could execute search to find the patterns of unknown events and determine learning laws from estimation of threats, predict and analyze threats, future prediction, construction of nested network of futuristic events. This module would be connected to the data acquisition. Futuristic analysis would be an integral part of every single element of the entire simulator.