Analog pattern categorization system having dual weighted connectivity between nodes

• US 5,179,596 A
• Filed: 07/05/1991
• Issued: 01/12/1993
• Est. Priority Date: 07/05/1991
• Status: Expired due to Fees

• Current Assignee: BOOZ Allen & Hamilton Incorporated (BOOZ Allen Hamilton Holding Corporation)
• Original Assignee: BOOZ Allen & Hamilton Incorporated (BOOZ Allen Hamilton Holding Corporation)
• Inventors: Weingard, Fred S.
• Primary Examiner(s): Boudreau, Leo H.
• Assistant Examiner(s): Fox, David
• Application Number:
  US07/726,052
• Time in Patent Office: 557 Days
• Field of Search: 382/14, 382/15, 364/274.9, 364/276.6, 364/972.4, 395/21, 395/23, 395/25
• US Class Current: 382/158
• CPC Class Codes: B82Y 30/00 Nanotechnology for materials or surface science, e.g. nanocomposites
• CPC Class Codes: G06F 18/2137 based on criteria of topology preservation, e.g. multidimensional scaling or self-organising maps
• CPC Class Codes: G06F 18/2433 Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
• CPC Class Codes: G06N 3/04 Architecture, e.g. interconnection topology

Abstract

Pattern categorization is provided by a self-organizing analog field/layer which learns many-to-many, analog spatiotemporal mappings. The field/layer employs a set of input nodes, each input node having two long term memory weights, and a set of output nodes. Each input node is for categorizing patterns with respect to a plurality of categories. The long term memory weights of an input node encode the patterns categorized by the input node. Each input node generates signals as a function of respective long term memory weights and input signals to the input node. Each input node is coupled to a different output node. Each output node receives signals generated by the respective input node and selects a category of the respective input node. The output nodes provide a mapping between plural parts of the input pattern and plural categories of the input nodes. Category selections of the output nodes are modified such that sum of the output signals from the output nodes is within a predefined range. Upon the sum of the output signals being within the predefined range, the output nodes provide categorization of the input pattern from the mapping between plural parts of the input pattern and plural categories of the input nodes.

Claims

1. A pattern categorization system comprising:

a short term memory presentation field for presenting input signals defining an input pattern;

a plurality of input nodes, each for categorizing patterns with respect to a plurality of categories, the input nodes being coupled to the presentation field for receiving the input signals, for each input signal received by an input node, the input node having two long term memory weights, the long term memory weights being indicative of a plurality of patterns categorized by the input node such that a net signal is generated as a function of the long term memory weights and input signals to the input node, and the net signal is indicative of similarity between the input pattern and patterns categorized by the input node;

a plurality of output nodes, one for each input node such that a different input node is connected to a different output node and each output node receives the net signal from the respective input node, and in response to the respective received net signal each output node selecting a category of the corresponding input node such that the output nodes provide a many-to-many mapping between plural parts of the input pattern and plural categories, each output node providing a short term memory output signal indicative of category selection; and

means for modifying category selections of the output nodes such that sum of the output signals from the output nodes is within a predefined range, upon the sum of the output signals being within the predefined range, the output nodes providing categorization of the input pattern from the mapping between plural parts of the input pattern and plural categories of the input nodes.

2. A system as claimed in claim 1 wherein for each input node one of the long term memory weights of the input node controls generalization of spatiotemporal parts of the input patterns.

3. A system as claimed in claim 1 wherein for each input node one of the long term memory weights of the input node controls strength of input signal to the input node, said strength being indicated as a degree of connectivity from the presentation field to the input node.

4. A system as claimed in claim 3 wherein the degree of connectivity is in the range of 0 to 1, 0 indicating absence of connection between the presentation field and the input node, and 1 indicating relative strength of connection between the presentation field and the input node.

5. A system as claimed in claim 1 wherein the predefined range is between 0 and the number of input nodes in the plurality of input nodes.

6. In a pattern recognition system, a method of categorizing patterns comprising the steps of:

in a short term memory presentation field, transmitting input signals defining an input pattern;

in each of a plurality of input nodes coupled to the presentation field, receiving the input signals and from the input signals determining a plurality of categorizing patterns with respect to a plurality of categories;

for each input signal received by an input node, applying respective two long term memory weights to the input signal, the long term memory weights being indicative of a plurality of patterns categorized by the input node;

for each input node, determining and generating a net signal from the long term memory weights of the input node and the input signals to the input node, the generated net signal being indicative of similarity between the input pattern and patterns categorized by the input node;

in a plurality of output nodes, one output node for each input node such that a different input node is connected to a different output node, each output node receiving the net signal from the respective input node, and in response, each output node selecting a category of the respective input node such that the output nodes provide a many-to-many mapping between plural parts of the input pattern and plural categories, each output node determining and transmitting a short term memory output signal indicative of category selection; and

modifying the category selections of the output nodes such that sum of the output signals transmitted from the output nodes is within a predefined range, upon the sum of the output signals being within the predefined range, the output nodes transmitting signals indicative of categorization of the input pattern from the many-to-many mapping between plural parts of the input pattern and plural categories of the input nodes.

7. A method as claimed in claim 6 wherein the step of applying two long term memory weights includes determining a first long term memory weight to control generalization of spatiotemporal parts of input patterns and a second long term memory weight to control strength of the input signal to the input node.

8. A method as claimed in claim 6 wherein the predefined range is between zero and the number of input nodes in the plurality of input nodes.

9. A pattern categorization system comprising:

a short term memory presentation field for presenting input signals defining an input pattern;

a plurality of input nodes, each for categorizing patterns with respect to a plurality of categories, the input nodes being coupled to the presentation field for receiving the input signals, for each input signal received by an input node, the input node having a first and second long term memory weight, the long term memory weights being indicative of a plurality of patterns categorized by the input node such that a net signal is generated as a function of the long term memory weights and input signals to the input node, and the net signal is indicative of similarity between the input pattern and patterns categorized by the input node;

for each input node, the first long term memory weight of the input node controls strength of the input signal to the input node, said strength being indicated as a degree of connectivity from the presentation field to the input node;

a plurality of output nodes, one for each input node such that a different input node is connected to a different output node and each output node receives the net signal form the respective input node, and in response to the respective received net signal, each output node selecting a category of the corresponding input node such that the output nodes provide a mapping between plural parts of the input pattern and plural categories, each output node providing a short term memory output signal indicative of category selection; and

means for modifying category selections of the output nodes such that sum of the output signals from the output nodes is within a predefined range, upon the sum of the output signals being within the predefined range, the output nodes providing categorization of the input pattern from the mapping between plural parts of the input pattern and plural categories of the input nodes.

10. A system as claimed in claim 9 wherein for each input node the second long term memory weight of the input node controls generalization of spatiotemporal parts of the input patterns.

11. A system as claimed in claim 9 wherein the degree of connectivity is in the range of 0 to 1, 0 indicating absence of connection between the presentation field and the input node, and 1 indicating relative strength of connection between the presentation field and the input node.

12. A system as claimed in claim 9 wherein the predefined range is between 0 and the number of input nodes in the plurality of input nodes.