System and method for a recursive cortical network

• US 9,607,263 B2
• Filed: 05/18/2016
• Issued: 03/28/2017
• Est. Priority Date: 05/15/2012
• Status: Active Grant

• Current Assignee: Intrinsic Innovation LLC (Alphabet Inc.), Intrinsic I. Llc
• Original Assignee: Vicarious FPC, Inc. (Alphabet Inc.)
• Inventors: George, Dileep, Kansky, Kenneth, Phoenix, D Scott, Marthi, Bhaskara, Laan, Christopher, Lehrach, Wolfgang
• Primary Examiner(s): Smith, Paulinho E
• Assistant Examiner(s): Chubb, Mikayla
• Application Number:
  US15/158,472
• Publication Number:
  US 20160260010A1
• Time in Patent Office: 314 Days
• US Class Current: 1/1
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/04 Architecture, e.g. interconnection topology
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06N 3/049 Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
• CPC Class Codes: G06N 3/082 modifying the architecture, e.g. adding, deleting or silencing nodes or connections
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A system includes: a recursively architected network of sub-networks organized into a hierarchical layers; the sub-networks including 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; the child feature node connectable to at least a parent feature node of a sub-network at a lower hierarchical layer.

Claims

1. A method for generating patterns with a network comprising:

providing a recursive network of sub-networks with a parent feature input node and at least two child feature output nodes;

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;

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

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 according to a lateral constraint node of the sub-network; and

compiling the state of final child feature nodes of the network into a generated output.

2. The method of claim 1, wherein the first node is the first PSCF node, and the second node of the second pool is a PSCF node.

3. The method of claim 2, wherein the second PSCF node of the second pool is within the same sub-network.

4. The method of claim 2, wherein the second PSCF node of the second pool is in a second sub-network.

5. The method of claim 2, wherein first PSCF node is in a network of a first time instant and the second PSCF node of the second pool is in a network of a second time instant.

6. The method of claim 2, further comprising in response to selecting the PSCF node selecting at least one child feature node;

mapping selected child feature nodes to data features that are compiled into the generated output.

7. The method of claim 6, wherein the generated output is an image.

8. The method of claim 6, wherein the generated output is a sequence of images.

9. The method of claim 6, wherein the generated output is audio.

10. The method of claim 6, wherein the selection functions are Boolean logical selection functions.

11. The method of claim 10, wherein the child feature nodes of a first layer sub-network are the parent feature nodes for at least two sub-networks in a second layer.

12. The method of claim 11, wherein selecting the pool node consistent with a selection function comprises selecting each pool node connected to the parent node according to a logical AND selection function;

wherein selecting the first PSCF node comprises randomly selecting only one PSCF node from the set of PSCF nodes connected to the pool node according to a logical XOR selection function of the pool node.