Application of hebbian and anti-hebbian learning to nanotechnology-based physical neural networks

• US 7,412,428 B2
• Filed: 12/30/2003
• Issued: 08/12/2008
• Est. Priority Date: 03/12/2002
• Status: Expired due to Term

• Current Assignee: Knowm Tech LLC
• Original Assignee: Knowm Tech LLC
• Inventors: Nugent, Alex
• Primary Examiner(s): Vincent; David R.
• Assistant Examiner(s): Tran; Mai T
• Application Number:
  US10/748,631
• Publication Number:
  US 20040162796A1
• Time in Patent Office: 1,687 Days
• Field of Search: 706/26, 706/6, 706/12, 706/15, 706/25, 977/700, 977/712, 977/720, 977/724, 977/742
• US Class Current: 706/26
• CPC Class Codes: B82Y 10/00 Nanotechnology for information processing, storage or transmission, e.g. quantum computing or single electron logic
• CPC Class Codes: G06N 3/065 Analogue means
• CPC Class Codes: G06N 3/088 Non-supervised learning, e.g. competitive learning
• CPC Class Codes: Y10S 977/70 Nanostructure
• CPC Class Codes: Y10S 977/712 formed from plural layers of nanosized material, e.g. stacked structures
• CPC Class Codes: Y10S 977/742 Carbon nanotubes, CNTs

Abstract

Methods and systems are disclosed herein in which a physical neural network can be configured utilizing nanotechnology. Such a physical neural network can comprise a plurality of molecular conductors (e.g., nanoconductors) which form neural connections between pre-synaptic and post-synaptic components of the physical neural network. Additionally, a learning mechanism can be applied for implementing Hebbian learning via the physical neural network. Such a learning mechanism can utilize a voltage gradient or voltage gradient dependencies to implement Hebbian and/or anti-Hebbian plasticity within the physical neural network. The learning mechanism can also utilize pre-synaptic and post-synaptic frequencies to provide Hebbian and/or anti-Hebbian learning within the physical neural network.

Claims

1. A system, comprising:

a physical neural network configured utilizing nanotechnology and integrated with feedback circuitry, wherein said physical neural network comprises a plurality of nanoconductors comprising at least one of nanotubes, nanowires, or nanoparticles, suspended and free to move about in a dielectric medium and which form neural connections between pre-synaptic and post-synaptic components of said physical neural network; and

a learning mechanism for applying Hebbian learning to said physical neural network.

2. The system of claim 1 wherein said learning mechanism utilizes a voltage gradient to implement Hebbian plasticity within said physical neural network.

3. The system of claim 1 wherein said learning mechanism utilizes voltage gradient dependencies associated with physical neural network to implement Hebbian learning within said physical neural network.

4. The system of claim 1 wherein said learning mechanism utilizes pre-synaptic and post-synaptic frequencies to provide Hebbian learning within said physical neural network.

5. The system of claim 1 wherein said learning mechanism utilizes a voltage gradient to implement anti-Hebbian plasticity within said physical neural network.

6. The system of claim 1 wherein said learning mechanism utilizes voltage gradient dependencies associated with physical neural network to implement anti-Hebbian learning within said physical neural network.

7. The system of claim 1 wherein said learning mechanism utilizes pre-synaptic and post-synaptic frequencies to provide anti-Hebbian learning within said physical neural network.

8. A system, comprising:

a physical neural network configured utilizing nanotechnology and integrated with feedback circuitry, wherein said physical neural network comprises a plurality of nanoconductors comprising at least one of nanotubes, nanowires, or nanoparticles, suspended and free to move about in a dielectric medium and which form neural connections between pre-synaptic and post-synaptic components of said physical neural network; and

a learning mechanism for applying Hebbian learning to said physical neural network wherein said learning mechanism utilizes a voltage gradient or pre-synaptic and post-synaptic frequencies thereof to implement Hebbian or anti-Hebbian plasticity within said physical neural network.

9. The system of claim 8 wherein said dielectric medium comprises a dielectric liquid.

10. The system of claim 9 wherein said plurality of nanoconductors form physical neural connections when said dielectric medium is exposed to an electric field, such that said physical neural connections can be strengthened or weakened depending upon a strengthening or weakening of said electric field or an alteration of a frequency thereof.