Neural network and method of neural network training

A method of training a neural network having an operative structure, the method comprising:

• receiving training images via a plurality of inputs to the neural network, wherein the training images are one of received as a training input value array and codified as the training input value array during training of the neural network;
• organizing corrective weights of a plurality of synapses in a corrective weight array, wherein each of the synapses is connected to one of the plurality of inputs and includes a plurality of corrective weights, and wherein each of the corrective weights is defined by a weight value;
• generating a neuron sum array via a plurality of neurons, wherein each of the neurons has at least one output and is connected with at least one of the plurality of inputs via at least one of the plurality of synapses, and wherein each respective neuron of the plurality of neurons is configured to add up the weight values of the corrective weights corresponding to each synapse connected to the respective neuron;
• receiving, via a controller, desired images organized as a desired output value array;
• determining, via the controller, a deviation of the neuron sum array from the desired output value array and generate a deviation array;
• modifying, via the controller, the corrective weight array using the determined deviation array, such that adding up the modified corrective weight values to determine the neuron sum array reduces the deviation of the neuron sum array from the desired output value array to generate a trained corrective weight array and thereby facilitate concurrent training of the neural network;
• and
• selectively arranging and rearranging connections between neural network elements, including the plurality of inputs, the plurality of neurons, and the plurality of synapses, during the training of the neural network via a structure-forming module;
• distributing the training images among the neural network elements via a signal allocation module during the training of the neural network;
• and
• commanding and coordinating operation of the controller, the structure-forming module, and the signal allocation module to reorganize the operative structure of the neural network during the training of the neural network via a training module, to thereby control, in-real time, the training of the neural network.

An operative structure of a neural network, the operative structure comprising:

• a plurality of inputs to the neural network configured to receive training images, wherein the training images are one of received as a training input value array and codified as the training input value array during training of the neural network;
• a plurality of synapses, wherein each synapse is connected to one of the plurality of inputs and includes a plurality of corrective weights, wherein each of the plurality of corrective weights is defined by a weight value, wherein the plurality of corrective weights of the plurality of synapses are organized in a corrective weight array, and wherein each synapse communicates weight signals indicative of weight values of the respective corrective weights;
• a plurality of neurons, wherein each neuron has at least one output and is connected with at least one of the plurality of inputs via at least one of the plurality of synapses, and wherein each of the plurality of neurons is configured to receive the weight signals from the at least one of the plurality of synapses and add up the weight values of the plurality of corrective weights corresponding to each of the plurality of synapses connected to the respective neuron, such that the plurality of neurons generate a neuron sum array;
• and
• a controller configured to:
• receive desired images organized as a desired output value array;
• determine a deviation of the neuron sum array from the desired output value array and generate a deviation array;
• and
• modify the corrective weight array using the determined deviation array, such that adding up the modified corrective weight values to determine the neuron sum array reduces the deviation of the neuron sum array from the desired output value array to generate a trained corrective weight array and thereby facilitate concurrent training of the neural network.