MULTI-MODAL NEURAL NETWORK FOR UNIVERSAL, ONLINE LEARNING

• US 20170185896A1
• Filed: 03/10/2017
• Published: 06/29/2017
• Est. Priority Date: 12/14/2011
• Status: Active Application

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Modha, Dharmendra S.
• Granted Patent: US 10,282,661 B2
• Time in Patent Office: Days
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: G06N 3/04 Architecture, e.g. interconnection topology
• CPC Class Codes: G06N 3/08 Learning methods

Abstract

In one embodiment, the present invention provides a neural network comprising multiple modalities. Each modality comprises multiple neurons. The neural network further comprises an interconnection lattice for cross-associating signaling between the neurons in different modalities. The interconnection lattice includes a plurality of perception neuron populations along a number of bottom-up signaling pathways, and a plurality of action neuron populations along a number of top-down signaling pathways. Each perception neuron along a bottom-up signaling pathway has a corresponding action neuron along a reciprocal top-down signaling pathway. An input neuron population configured to receive sensory input drives perception neurons along a number of bottom-up signaling pathways. A first set of perception neurons along bottom-up signaling pathways drive a first set of action neurons along top-down signaling pathways. Action neurons along a number of top-down signaling pathways drive an output neuron population configured to generate motor output.

Claims

1. A method comprising:

receiving motor output from one or more neurons of a neural network in response to the one or more neurons firing;

determining a type of the motor output; and

based on the type of the motor output, selectively routing the motor output to one or more other neurons of the neural network for training.

2. The method of claim 1, wherein the type of the motor output is one of a first type or a second type.

3. The method of claim 1, wherein selectively routing the motor output comprises:

routing the motor output to the one or more other neurons of the neural network for learning in response to determining the type of the motor output is the first type.

4. The method of claim 3, wherein the first type represents a false negative.

5. The method of claim 1, wherein selectively routing the motor output comprises:

routing the motor output to the one or more other neurons of the neural network for unlearning in response to determining the type of the motor output is the second type.

6. The method of claim 5, wherein the second type represents a false positive.

7. The method of claim 1, wherein the neural network comprises a plurality of neurons interconnected via a first set of signaling pathways and a second set of signaling pathways.

8. The method of claim 7, further comprising:

training each neuron along a signaling pathway of the first set using a first learning rule; and

training each neuron along a signaling pathway of the second set using a second learning rule different from the first learning rule.

9. A system comprising a computer processor, a computer-readable hardware storage device, and program code embodied with the computer-readable hardware storage device for execution by the computer processor to implement a method comprising:

receiving motor output from one or more neurons of a neural network in response to the one or more neurons firing;

determining a type of the motor output; and

based on the type of the motor output, selectively routing the motor output to one or more other neurons of the neural network for training.

10. The system of claim 9, wherein the type of the motor output is one of a first type or a second type.

11. The system of claim 9, wherein selectively routing the motor output comprises:

routing the motor output to the one or more other neurons of the neural network for learning in response to determining the type of the motor output is the first type.

12. The system of claim 11, wherein the first type represents a false negative.

13. The system of claim 9, wherein selectively routing the motor output comprises:

routing the motor output to the one or more other neurons of the neural network for unlearning in response to determining the type of the motor output is the second type.

14. The system of claim 13, wherein the second type represents a false positive.

15. The system of claim 9, wherein the neural network comprises a plurality of neurons interconnected via a first set of signaling pathways and a second set of signaling pathways.

16. The system of claim 15, the method further comprising:

training each neuron along a signaling pathway of the first set using a first learning rule; and

training each neuron along a signaling pathway of the second set using a second learning rule different from the first learning rule.

17. A computer program product comprising a computer-readable hardware storage device having program code embodied therewith, the program code being executable by a computer to implement a method comprising:

receiving motor output from one or more neurons of a neural network in response to the one or more neurons firing;

determining a type of the motor output; and

based on the type of the motor output, selectively routing the motor output to one or more other neurons of the neural network for training.

18. The computer program product of claim 17, wherein selectively routing the motor output comprises:

routing the motor output to the one or more other neurons of the neural network for learning in response to determining the type of the motor output is a false negative.

19. The computer program product of claim 17, wherein selectively routing the motor output comprises:

routing the motor output to the one or more other neurons of the neural network for unlearning in response to determining the type of the motor output is a false positive.

20. The computer program product of claim 17, wherein the neural network comprises a plurality of neurons interconnected via a first set of signaling pathways and a second set of signaling pathways, each neuron along a signaling pathway of the first set is trained using a first learning rule, and each neuron along a signaling pathway of the second set is trained using a second learning rule different from the first learning rule.