Method for substituting an electronic emulation of the human brain into an application to replace a human

• US 8,001,067 B2
• Filed: 04/07/2007
• Issued: 08/16/2011
• Est. Priority Date: 01/06/2004
• Status: Expired due to Fees

• Current Assignee: Neuric Technologies, LLC
• Original Assignee: Neuric Technologies, LLC
• Inventors: Fettig, Matthew, Divar, Vijay, Womack, Lukas K., Hamilton, Gene P., Visel, Thomas A.
• Primary Examiner(s): Gaffin; Jeffrey A
• Assistant Examiner(s): Kim; David H
• Application Number:
  US11/697,721
• Publication Number:
  US 20070282765A1
• Time in Patent Office: 1,592 Days
• Field of Search: 706/46, 706/23
• US Class Current: 706/23
• CPC Class Codes: G06N 3/10 Interfaces, programming languages or software development kits, e.g. for simulating neural networks

Abstract

A method for emulating human behavior and actions in an expert mode to control and execute a defined task. The first step is to define a task based application having a task associated therewith that is operable to receive goals and conditions and which goals define the desired results to be achieved when the task is performed. A brain emulation is represented by a plurality of nodes each representing a concept, and interconnecting relationships between select ones of the concepts, which brain emulation is operable to receive information from the task based application and process such information and capable controlling the task based application with the outcome of such processing to achieve the goal of the task based application. The brain emulation receives information either directly from a external source or from the task based application as to the state of the task based application and processes the received information based on the existing interconnecting relationships to initiate the task associated with the task based application and attempt to achieve the goal associated with the task based application.

Claims

1. A method for emulating human behavior and actions to control and execute a defined task, comprising the steps of:

defining a task based application having a task associated therewith that is operable to receive goals and conditions and which goals define desired results to be achieved when the task is performed;

providing inputs to the task based application for controlling the operation thereof;

receiving from the task based application information as to the state of the task based application;

defining a brain emulation that is represented by a plurality of concept nodes each representing a concept, and non-fixed interconnecting relationships defined between select ones of the concept nodes, which brain emulation is operable to receive information from the task based application and process such information and capable of controlling the task based application with the outcome of such processing to achieve the goal of the task based application;

training, using a processor, the brain emulation in a training mode to create and define the otherwise non pre-existing relationships between concepts associated with the concept nodes in response to training information input thereto that is designed to impart predetermined meaning to the one or more concepts in the brain emulation and wherein some of the concept nodes have bidirectional relationships;

operating in an operational mode after training to receive information either directly from an external source or from the task based application as to the state of the task based application; and

process the received information based on the existing interconnecting relationships to initiate the task associated with the task based application and attempt to achieve the goal associated with the task based application.

2. The method of claim 1, wherein received information from the task based application includes information about a surrounding environment of the task based application during the performance of the associated task and changes thereto.

3. The method of claim 1, wherein the task is a temporal task that occurs over a time horizon.

4. The method of claim 1, wherein the step of training comprises the steps of:

training the brain emulation with a first generic set of general information to establish general relationships between concepts in response to general training information input thereto that is designed to impart a general predetermined meaning to the one or more concepts in the brain emulation that are not necessarily specific to the task based application; and

training the brain emulation with a second and task based application specific set of information to establish task specific relationships between concepts associated with the task based application in response to task specific training information focused on the associated task and input thereto that is designed to impart a specific predetermined task related meaning to the one or more concepts in the brain emulation that are specific to the task based application.

5. The method of claim 1, wherein, in response to external disturbances having no predetermined or trained relationships associated therewith, the external disturbances will be received and used to update the existing trained relationships as experiences that can be used in a future performance of the task.

6. The method of claim 1, where in the goals of the task based application are imparted to the brain emulation after the step of training, such that the brain emulation has the training thereof and the learned relationships modified.