System and method for a recursive cortical network

1. A method for inferring patterns in image data, the data having a space of two or more dimensions, with a recursive neural network comprising:

providing a recursive network of sub-networks with a parent feature node and at least two child feature nodes; wherein each sub-network is associated with a distinct subset of the space;

configuring nodes of the sub-networks with posterior distribution component;

enforcing a selection constraint between at least two nodes of the network;

receiving image data feature input at the final child feature nodes;

propagating node activation through the network layer hierarchy in a manner consistent with node connections of sub-networks of the network and the posterior prediction of child nodes; and

outputting parent feature node selection to an inferred output, wherein the inferred output is representative of a known pattern.

2. The method of claim 1 , wherein the distinct subsets of the space are non-overlapping.

3. The method of claim 1 , wherein the selection constraint is enforced by a lateral constraint node between the two nodes; wherein the two nodes are in a same sub-network.

4. The method of claim 1 , wherein the selection constraint is enforced by an external constraint node between the two nodes; wherein the two nodes are in different sub-networks.

5. The method of claim 1 , wherein the network contains at least two sets of sub-networks; the first set of sub-networks associated with a first time and the second set of the sub-networks associated with a second time; wherein the selection constraint is enforced by a temporal constraint node between the two nodes; wherein the two nodes are in sub-networks of different sets.

6. A method for generating image data, the data having a space of two or more dimensions, with a recursive neural network comprising:

providing a recursive network of sub-networks with a parent feature input node and at least two child feature output nodes; wherein each sub-network is associated with a distinct subset of the space;

propagating node selection down the network layer hierarchy in a manner consistent with node connections of sub-networks of the network, the propagation within the sub-network comprising:

at an activated parent feature node of a sub-network, selecting a pool node consistent with a selection function of the parent feature node; and

at a pool node of a sub-network, selecting at least a first parent-specific child feature (PSCF) node that corresponds to a child feature node of the sub-network; and

generating an image data output, comprising compiling the state of final child feature nodes of the network into the generated image data output.

7. The method of claim 6 , wherein the distinct subsets of the space are non-overlapping.

8. The method of claim 6 , further comprising in response to the selection of a first node of the sub-network, enforcing a selection constraint on at least a second node of a second pool.

9. The method of claim 8 , wherein the selection constraint is enforced by a lateral constraint node between the first and second nodes; wherein the first and second nodes are in a same sub-network.

10. The method of claim 8 , wherein the selection constraint is enforced by a lateral constraint node between the first and second nodes; wherein the first and second nodes are in different sub-networks.

11. The method of claim 8 , wherein the network contains at least two sets of sub-networks; the first set of sub-networks associated with a first time and the second set of the sub-networks associated with a second time; wherein the selection constraint is enforced by a temporal constraint node between the first and second nodes; wherein the first and second nodes are in sub-networks of different sets.

12. A system comprising a computer processor configured to receive a first image depicting an object and a second image depicting a transformed object, the system configured to run:

a recursively architected network of sub-networks organized into a plurality of hierarchical layers;

the sub-networks comprising at least a parent feature node, a pool node, a parent-specific child feature (PSCF) node, and a child feature node;

the parent feature node of at least one sub-network configured with a selection function actionable on at least two pool nodes connected to the parent feature node of the at least one sub-network;

the pool node of the at least one sub-network configured with a selection function actionable on at least two PSCF nodes connected to the pool node of the at least one sub-network;

the PSCF node of the at least one sub-network configured to activate a connected child feature node; and

the child feature node connectable to at least a parent feature node of a sub-network at a lower hierarchical layer;

wherein the recursively architected network is configured to identify the transformed object depicted in the second image as the object depicted in the first image.

13. The system of claim 12 , further comprising a constraint node with at least two connections from at least two PSCF nodes, with a selection function to augment selection by the pool node.

14. The system of claim 13 , wherein the constraint node is a lateral constraint node between two PSCF nodes of a same sub-network.

15. The system of claim 13 , wherein the constraint node is an external constraint node between two PSCF nodes of different sub-networks.

16. The system of claim 13 , wherein the constraint node is a temporal constraint node between two PSCF nodes of a different sub-networks; the different sub-networks corresponding to different temporal ranges but to a same spatial range.

17. The system of claim 12 ,

wherein the child feature node is configured to process image information, associated with the second image, into child image information;

wherein the pool node is configured to pass the child image information to the parent feature node when the PSCF node, associated with the child feature node, is activated; and

wherein the parent feature node is configured to facilitate identification of the transformed object within the second image as the object depicted in the first image based on the child image information and auxiliary child image information from the other pool node of the at least two pool nodes.