NEUROMORPHIC SPATIOTEMPORAL WHERE-WHAT MACHINES

• US 20100312730A1
• Filed: 05/28/2010
• Published: 12/09/2010
• Est. Priority Date: 05/29/2009
• Status: Active Grant

• Current Assignee: The Board of Trustees of Michigan State University (Michigan State University)
• Original Assignee: The Board of Trustees of Michigan State University (Michigan State University)
• Inventors: Ji, Zhengping, Weng, Juvang, Solgi, Mojtaba, Luciw, Matthew
• Granted Patent: US 8,694,449 B2
• Time in Patent Office: Days
• US Class Current: 706/15
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: G06N 3/08 Learning methods

Abstract

In various embodiments, electronic apparatus, systems, and methods include a unified compact spatiotemporal method that provides a process for machines to deal with space and time and to deal with sensors and effectors. Additional apparatus, systems, and methods are disclosed.

Claims

1. A method comprising:

generating, within a machine using a processing unit, actions for the machine based on correlating spatial inputs linked in time for each neuron in a set of neurons on a neuronal layer for the machine such that correlation of the correlating spatial inputs includes applying a firing-age dependent plasticity, wherein the spatial inputs are from an input layer and a corresponding output layer.

2. The method of claim 1, wherein the method includes computing responses for all the neurons, the response for each neuron based on relevance of sensor inputs and effector inputs for the respective neuron.

3. The method of claim 2, wherein computing responses for all the neurons includes computing responses for all the neurons at a time of t+τ

, τ

>

0, in which responses, only at time t, are included in computing the responses at t+τ

.

4. The method of claim 1, wherein generating autonomous decisions results from an iterative selection of a neuron from the set of neurons best matched to a feature from a sensory input relative to a motor input.

5. The method of claim 1, wherein the method includes operating on the neurons without using error signals for back-propagation.

6. A machine-readable medium that stores instructions, which when performed by a machine, cause the machine to perform operations comprising:

generating, within the machine using a processing unit, autonomous decisions for the machine based on correlating spatial inputs linked in time for each neuron in a set of neurons on a neuronal layer for the machine such that correlation of the correlating spatial inputs includes applying a firing-age dependent plasticity, the spatial inputs being from an input layer and from a corresponding output layer.

7. The machine-readable medium of claim 6, wherein the instructions include computing responses for all the neurons, the response for each neuron based on relevance of sensor inputs and effector inputs for the respective neuron.

8. The machine-readable medium of claim 7, wherein computing responses for all the neurons include computing responses for all the neurons at a time of t+τ

, τ

>

0, in which responses, only at time t, are included in computing the responses at t+τ

.

9. The machine-readable medium of claim 6, wherein generating autonomous decisions results from an iterative selection of a neuron from the set of neurons best matched to a feature from a sensory input relative to a motor input.

10. The machine-readable medium of claim 6, wherein the instructions include operating on the neurons without using error signals for back-propagation.

11. A method comprising:

extracting features from a sensory input provided by a sensory device for a machine;

correlating motor action of the machine with the features; and

generating autonomously, using a processing unit, a decision for machine activity based on results from correlating motor action with the features based on applying a firing-age dependent plasticity for neurons associated with the sensory input and motor action.

12. The method of claim 11, wherein the method further comprises:

determining whether one or more of the features are irrelevant to motor actions; and

automatically disregarding features determined to be irrelevant.

13. The method of claim 11, wherein generating decisions includes operating in a neural network mode without error back-propagation.

14. The method of claim 11, wherein the method includes correlating motor action of the machine with the features relative to a fixed time period such that update of information relative to correlating motor action of the machine with the features at a current time is based only on information corresponding to motor action of the machine with the features at a previous time set by the fixed time period.

15. The method of claim 11, wherein the method further comprises:

extracting an image from the sensory input; and

generating a decision based on recognizing an object in the image separate from background in the image.

16. A system comprising:

a sensory device;

a motor device;

memory; and

a processing unit coupled to the sensory device, to the motor device, and to the memory, the processing unit arranged to operatively execute instructions stored in the memory, which when executed by the processing unit, cause the system to perform operations including generating, within the system using the processing unit, autonomous decisions for the system based on correlating spatial inputs linked in time for each neuron in a set of neurons on a neuronal layer for the system such that the correlation of the spatial inputs includes applying a firing-age dependent plasticity, the spatial inputs being from an input layer and from a corresponding output layer.

17. The system of claim 16, wherein the processing unit and the memory are operable to compute responses for all the neurons includes computing responses for all the neurons at a time of t+τ

, τ

>

0, in which responses, only at time t, are included in computing the responses at t+τ

.

18. The system of claim 16, wherein the processing unit and the memory are operable to determine whether one or more of features extracted from the sensory input are irrelevant to motor actions of the system and to automatically disregard features determined to be irrelevant.

19. The system of claim 16, wherein the system is arranged as a distributed system.

20. The system of claim 16, wherein the system is arranged as a robot-like apparatus.