SEQUENCE LEARNING IN A HIERARCHICAL TEMPORAL MEMORY BASED SYSTEM

• US 20100049677A1
• Filed: 10/09/2009
• Published: 02/25/2010
• Est. Priority Date: 02/10/2006
• Status: Active Grant

• Current Assignee: Numenta, Inc.
• Original Assignee: Numenta, Inc.
• Inventors: Hawkins, Jeffrey, George, Dileep, Astier, Frank, Jaros, Robert G.
• Granted Patent: US 8,285,667 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/04 Architecture, e.g. interconnection topology
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06N 3/047 Probabilistic or stochastic networks
• CPC Class Codes: G06N 3/049 Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
• CPC Class Codes: G06N 3/08 Learning methods
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A hierarchy of computing modules is configured to learn a cause of input data sensed over space and time, and is further configured to determine a cause of novel sensed input data dependent on the learned cause. At least one of the computing modules has a sequence learner module configured to associate sequences of input data received by the computing module to a set of causes previously learned in the hierarchy.

Claims

1. A computer-implemented system, comprising:

a hierarchy of computing modules configured to receive first input data to learn spatial patterns and temporal sequences in the first input data associated with an object or a state of the object in a learning stage, the hierarchy in an inference stage subsequent to the learning stage further configured receive second input data and generate the output information representing probabilities that spatial patterns and temporal sequences in the second input data correspond to spatial patterns and temporal sequences learned in the learning stage, wherein at least one of the computing modules comprises;

a sequence learner module in the learning stage configured to associate temporal sequences of spatial patterns in the first input data with the output information, the associated temporal sequences having different sequence lengths, the at least one computing module generating the output information responsive to receiving the second input data based on the association in the inference stage.

2. The system of claim 1, wherein the at least one computing module further comprises:

a coincidence detection module configured to detect coincidences among input data received by the computing module, the coincidence detector module further configured to send information about spatial patterns in the first input data corresponding to the coincidences to the sequence learner module.

3. The system of claim 1, wherein the output information comprises a distribution representing probabilities of the second input data caused by an object or a state of an object in the inference stage, the distribution being dependent on the association of temporal sequences with the output information generated in the learning stage.

4. The system of claim 3, wherein the sequence learner has a predetermined number of variables in the output information.

5. The system of claim 3, wherein the sequence learner module is further configured to adjust the distribution as temporal sequences of spatial patterns in the second input data changes over time in the inference stage.

6. The system of claim 1, wherein at least two of the temporal sequences share one or more spatial patterns.

7. A computer-implemented method, comprising:

receiving spatial patterns in first input data in a learning stage;

identifying temporal sequences of the spatial patterns that occur with frequency above a predetermined statistical threshold in the learning stage, the identified temporal sequences having different sequence lengths;

associating the temporal sequences with the output information based on the identified temporal sequences in the learning stage; and

outputting, in an inference stage subsequent to the learning stage, the output information responsive to receiving temporal sequences of spatial patterns in second input data based on the association.

8. The computer-implemented method of claim 7, wherein the output information comprises a distribution representing probabilities that the temporal sequences are caused by an object or a state of an object.

9. The computer-implemented method of claim 8, further comprising:

adjusting the distribution as received spatial patterns in the second input data change over time.

10. The computer-implemented method of claim 7, wherein at least two of the temporal sequences share one or more spatial patterns.

11. A computer-implemented method providing output information for determining an object or a state of an object associated with spatial patterns and temporal sequences in input data, comprising:

receiving spatial patterns in first input data in a learning stage;

storing a count of how many times a particular temporal sequence has been received in the first input data during the learning stage;

identifying temporal sequences of the spatial patterns that occur with frequency above a predetermined statistical threshold based on the stored count in the learning stage;

associating the temporal sequences with the output information in the learning stage based on the identified temporal sequences in the learning stage; and

outputting, in an inference stage subsequent to the learning stage, the output information responsive to receiving temporal sequences of spatial patterns in second input data based on the association.

12. The computer-implemented method of claim 11, wherein the count is stored dependent on whether the particular temporal sequence has been received at a statistically frequent rate.

13. The computer-implemented method of claim 11 wherein the received temporal sequence is associated with the output information dependent on the temporal sequence being identified as being received at a statistically frequent rate.

14. The computer-implemented method of claim 11, wherein at least two of the temporal sequences share one or more spatial patterns.

15. A computer-implemented method providing output information for determining an object or a state of an object associated with spatial patterns and temporal sequences in input data, comprising:

receiving spatial patterns in first input data in a learning stage;

collecting data on received temporal sequences of the spatial patterns that are shorter than a predetermined sequence length;

identifying temporal sequences shorter than the predetermined sequence length that occur with frequency above a predetermined statistical threshold in the learning stage, at least two of the identified temporal sequences sharing one or more spatial patterns;

associating the temporal sequences with the output information in the learning stage based on the identified temporal sequences; and

outputting, in an inference stage subsequent to the learning stage, the output information responsive to receiving temporal sequences of spatial patterns in second input data based on the association.

16. The computer-implemented method of claim 15, wherein the data is collected on the received temporal sequences of a first length that have been identified as being statistically frequent based on receiving temporal sequences of a second lengths that is shorter than the first length.

17. A computer-readable medium having instructions stored therein that are executable on a processor, the instructions configured to provide output information for determining an object or a state of an object associated with spatial patterns and temporal sequences in input data, the instructions when executed cause the processing system to:

receive spatial patterns in first input data in a learning stage;

identify temporal sequences of the spatial patterns that occur with frequency above a predetermined statistical threshold in the learning stage, the identified temporal sequences having different sequence lengths;

associate the temporal sequences with the output information in the learning stage based on the identified temporal sequences in the learning stage; and

output, in an inference stage subsequent to the learning stage, the output information responsive to receiving temporal sequences of spatial patterns in second input data based on the association.

18. The computer-readable medium of claim 17, wherein the output information comprises a distribution representing probabilities that the temporal sequences are caused by an object or a state of an object.

19. The computer-readable medium of claim 18, further comprising instructions to:

adjust the distribution as received spatial patterns in the second input data change over time.

20. The computer-readable medium of claim 17, wherein at least two of the temporal sequences share one or more spatial patterns.

21. A computer-readable medium having instructions stored therein that are executable on a processor, the instructions configured to provide output information for determining an object or a state of an object associated with spatial patterns and temporal sequences in input data, the instructions when executed cause the processing system to:

receive spatial patterns in first input data in a learning stage;

store count of how many times a particular temporal sequence has been received in the learning stage;

identify temporal sequences of the spatial patterns that occur with frequency above a predetermined statistical threshold in the learning stage;

associate the temporal sequences with the output information in the learning stage based on the identified temporal sequences in the learning stage; and

output, in an inference stage subsequent to the learning stage, the output information responsive to receiving temporal sequences of spatial patterns in second input data based on the association.

22. The computer-readable medium of claim 21, wherein the count is stored dependent on whether the particular temporal sequence has been received at a statistically frequent rate.

23. The computer-readable medium of claim 21, wherein a received temporal sequence is associated with the output information dependent on the temporal sequence being identified as being received at a statistically frequent rate.

24. A computer-readable medium having instructions stored therein that are executable on a processor, the instructions configured to provide output information for determining an object or a state of an object associated with spatial patterns and temporal sequences in input data, the instructions when executed cause the processing system to:

receive spatial patterns in first input data in a learning stage;

collect data on received temporal sequences of the spatial patterns that are shorter than a predetermined sequence length;

identify the temporal sequences that occur with frequency above a predetermined statistical threshold in the learning stage;

associate the temporal sequences with the output information in the learning stage based on the identified temporal sequences in the learning stage; and

output, in an inference stage subsequent to the learning stage, the output information responsive to receiving temporal sequences of spatial patterns in second input data based on the association.

25. The computer-readable medium of claim 24, further comprising instructions to:

collect data on received temporal sequences that have been identified as being statistically frequent based on observations of shorter length temporal sequences.