Coupling multiple artificially learning units with a projection level

1 . A method in a system of at least a first artificial intelligence unit, second artificial intelligence unit, and a third artificial intelligence unit comprising:

inputting one or more first input values to the first artificial intelligence unit to obtain one or more first output values, wherein the one or more first input values comprise at least one of sensor data, audio data, or image data;

forming a modulation function based on the one or more first output values of the first artificial intelligence unit or one or more third output values of the third artificial intelligence unit;

applying the modulation function to one or more parameters of the second artificial intelligence unit, wherein the second artificial intelligence unit comprises a larger number of nodes than the first artificial intelligence unit, and the one or more parameters comprise at least one of: a weighting for a node of the second artificial intelligence unit, an activation function of the node, an output function of the node, or a propagation function of the node;

inputting the one or more first input values to the second artificial intelligence unit to obtain one or more second output values;

at least temporarily storing situation data comprising at least one of the one or more first input values or the one or more second output values;

using the situation data as one or more third input values for the third artificial intelligence unit to obtain the one or more third output values; and

checking whether the one or more second output values satisfy one or more predetermined conditions based on the one or more third output values, wherein at least one of the one or more first output values or the one or more second output values is used as one or more direct or indirect control signals for an actuator of an automated system, wherein (i) after the expiration of a first time period, the one or more first output values are used, or (ii) after the expiration of a second time period, which starts with the expiration of the first time period, the one or more second output values are used.

2 . The method of claim 1 , wherein the one or more second output values are discarded if the one or more second output values do not satisfy at least one of the one or more predetermined conditions.

3 . The method of claim 1 , further comprising:

deciding whether at least a portion of the situation data should be stored permanently; and p 1 transferring the situation data to be stored permanently to a long-term memory.

4 . The method of claim 3 , wherein deciding whether to permanently store at least the portion of the situation data is dependent on whether the one or more second output values satisfy the one or more predetermined conditions.

5 . The method of claim 1 , wherein storing the situation data further comprises storing timing information for the one or more first input values or the one or more second output values.

6 . The method of claim 5 , wherein the timing information comprises one or more time intervals associated with the one or more first input values or the one or more second output values.

7 . The method of claim 1 , wherein the situation data is temporarily stored for at least a predetermined period of time, and wherein at least one of the first artificial intelligence unit, the second artificial intelligence unit, or the third artificial intelligence unit is adapted to set or change the predetermined period of time.

8 . The method of claim 1 , comprising comparing the one or more second output values with the situation data.

9 . The method of claim 1 , wherein;

(i) each of the first artificial intelligence unit, the second artificial intelligence unit, and the third artificial intelligence unit is assigned a respective classification memory,

ii) each of the first artificial intelligence unit, the second artificial intelligence unit, and the third artificial intelligence unit, when generating one or more output values, classifies one or more input values into one or more classes (K1, . . . Kj, Km, Kn) which are stored in the respective classification memory, the one or more classes each being structured in one or more dependent levels, and

iii) the second artificial intelligence unit is associated with a larger number of classes or a larger number of depending levels than the first artificial intelligence unit or the third artificial intelligence unit.

10 . The method of claim 1 , wherein one or more of the first artificial intelligence unit, the second artificial intelligence unit, and the third artificial intelligence unit comprise a neural network.

11 . A system comprising:

at least a first artificial intelligence unit, second artificial intelligence unit, and a third artificial intelligence unit;

means for acquiring one or more first input values; and

at least one user interface for outputting one or more total output values to a user, wherein the one or more total output values are formed based on one or more output values from one or more of the first artificial intelligence unit, the second artificial intelligence unit, and the third artificial intelligence unit by:

inputting the one or more first input values to the first artificial intelligence unit to obtain one or more first output values, wherein the one or more first input values comprise at least one of sensor data, audio data, or image data;

forming a modulation function based on the one or more first output values of the first artificial intelligence unit or one or more third output values of the third artificial intelligence unit;

applying the modulation function to one or more parameters of the second artificial intelligence unit, wherein the second artificial intelligence unit comprises a larger number of nodes than the first artificial intelligence unit, and the one or more parameters comprise at least one of: a weighting for a node of the second artificial intelligence unit, an activation function of the node, an output function of the node, or a propagation function of the node;

inputting the one or more first input values to the second artificial intelligence unit to obtain one or more second output values;

at least temporarily storing situation data comprising at least one of the one or more first input values or the one or more second output values;

using the situation data as one or more third input values for the third artificial intelligence unit to obtain the one or more third output values; and

checking whether the one or more second output values satisfy one or more predetermined conditions based on the one or more third output values, wherein at least one of the one or more first output values or the one or more second output values is used as one or more direct or indirect control signals for an actuator of an automated system, wherein (i) after the expiration of a first time period, the one or more first output values are used, or (ii) after the expiration of a second time period, which starts with the expiration of the first time period, the one or more second output values are used.

12 . The system of claim 11 , further comprising:

a projection plane adapted to at least temporarily store the situation data.

13 . The system of claim 12 , further comprising at least one sensor element or at least one detection unit, wherein the one or more first input values comprise one or more values detected by the at least one sensor element or the at least one detection unit.

14 . The system of claim 12 , further comprising a storage element of the projection plane in which the situation data is stored.

15 . The system of claim 12 , further comprising at least one output module for outputting the one or more total output values to the user, wherein the at least one output module comprises at least one of: a display screen, a touch screen, a speaker, or a projection module.

16 . The system of claim 12 , wherein each of the first artificial intelligence unit, the second artificial intelligence unit, and the third artificial intelligence unit is assigned a respective classification memory, each of the first artificial intelligence unit, the second artificial intelligence unit, and the third artificial intelligence unit is arranged to perform a classification of one or more input values into one or more classes (K1, . . . Kj, Km, Kn) which are stored in the respective classification memory, the one or more classes each being structured in one or more dependent levels, and the second artificial intelligence unit is associated with a larger number of classes or a larger number of depending levels than the first artificial intelligence unit or the third artificial intelligence unit.

17 . The system of claim 12 , wherein one or more of the first artificial intelligence unit, the second artificial intelligence unit, and the third artificial intelligence unit comprise a neural network.