Input into a neural network

1. A recurrent artificial neural network comprising:

a first region of the recurrent artificial neural network that is configured to receive and process data originating from a first sensor;

a second region of the recurrent artificial neural network that is configured to receive and process data originating from a second sensor, wherein the first sensor differs from the second sensor, wherein the first region is primarily perturbed by data originating from the first sensor and the second region is primarily perturbed by data originating from the second sensor even when both regions are perturbed at the same time; and

a third region of the recurrent artificial neural network that is configured to receive results of the processing by both the first region and by the second regions, wherein the third region is configured to output indications of the presence or absence of topological patterns of signal transmission activity within the recurrent artificial neural network, wherein the topological patterns are responsive to the results of the processing by the first region and by the second regions.

2. The recurrent artificial neural network of claim 1 , wherein the third regions is configured to perform operations comprising:

identifying topological patterns of activity in the recurrent artificial neural network, wherein the topological patterns are responsive to input, into the recurrent artificial neural network, of first data originating from the first sensor and second data originating from the second sensor and each topological pattern abstracts a characteristic shared by the first data and the second data, wherein the first and second sensors sense different data; and

outputting a collection of digits, wherein each digit represents whether one of the topological patterns of activity has been identified in the artificial neural network.

3. The recurrent artificial neural of claim 2 , wherein the operations further comprise defining a plurality of windows of time during which the activity of the artificial neural network is responsive to an input into the artificial neural network, wherein the topological patterns of activity are identified in each of the pluralities of windows of time.

4. The recurrent artificial neural of claim 2 , wherein the first sensor produces a stream of output data and the second sensor produces slower changing or static output data.

5. The recurrent artificial neural of claim 4 , wherein the slower changing or static output data is rate coded and the rate coded data is input into the recurrent artificial neural network at a same time as when the data that originates from the first transducer is input into the recurrent artificial neural network.

6. The recurrent artificial neural of claim 2 , wherein the digits are multi-valued and represent a probability that the topological pattern of activity is present in the artificial neural network.

7. The recurrent artificial neural of claim 2 , wherein:

the recurrent artificial neural further comprises a third region that is configured to receive third data originating from a third sensor, wherein the third sensor senses data that differs from the first and second data; and

the third region is configured to identify topological patterns of signal transmission activity within the recurrent artificial neural network that abstract a characteristic shared by the first data, the second data, and the third data.

8. The recurrent artificial neural network of claim 2 , wherein the topological patterns of activity are clique patterns.

9. The recurrent artificial neural network of claim 8 , wherein the clique patterns of activity enclose cavities.

10. The recurrent artificial neural network of claim 1 , wherein each of the first region and the second region is an identifiably discrete collection of nodes and edges with relatively few node-to-node connections between the first region and the second region.

11. The recurrent artificial neural network of claim 1 , wherein the first region is configured to output indications of the presence of topological patterns of signal transmission activity within the first region that are primarily responsive to the input of the data originating from the first sensor.

12. The recurrent artificial neural network of claim 1 , wherein the topological patterns of signal transmission activity are clique patterns.

13. The recurrent artificial neural network of claim 12 , wherein the clique patterns of signal transmission activity enclose cavities.

14. The recurrent artificial neural network of claim 12 , wherein the clique patterns are directed clique patterns.

15. The recurrent artificial neural network of claim 1 , wherein the first sensor produces a stream of output data and the second sensor produces slower changing or static output data.

16. The recurrent artificial neural network of claim 15 , further comprising a rate coder coupled to rate code the slower changing or static output data and input the rate coded data into the second region at a same time as when the data that originates from the first sensor is input into the first region.

17. The recurrent artificial neural network of claim 1 , further comprising means for scaling a magnitude of the data originating from the first sensor prior to receipt by the first region, wherein the scaling is based on the data originating from the second sensor.

18. The recurrent artificial neural network of claim 1 , further comprising an input coupled to inject some of the data originating from the first sensor into a node or link of the recurrent neural network, wherein the input includes a delay or a scaling element, wherein a magnitude of the delay or of the scaling is based on the data originating from the second sensor.