Interface neural network

1. An operation method of an interface neural network, the operation method comprising:

receiving an output signal from a first neural network;

converting a first feature vector of the output signal to a second feature vector of an input signal to be input to a second neural network, based on parameters of the interface neural network controlling conversion between a feature vector to be output from the first neural network and a feature vector to be input to the second neural network;

generating the input signal to be input to the second neural network, the input signal comprising the second feature vector; and

transmitting the input signal to the second neural network.

2. The operation method of claim 1 , wherein an input dimension of the interface neural network corresponds to an output dimension of the first neural network, and

an output dimension of the interface neural network corresponds to an input dimension of the second neural network.

3. The operation method of claim 1 , wherein the conversion the parameters of the interface neural network are optimized.

4. The operation method of claim 1 , further comprising:

based on the first neural network being replaced with a third neural network, updating the parameters to control conversion between a feature vector to be output from the third neural network and the feature vector to be input to the second neural network; and

based on the second neural network being replaced with a fourth neural network, updating the parameters to control conversion between the feature vector to be output from the first neural network and a feature vector to be input to the fourth neural network.

5. The operation method of claim 4 , wherein the updating the parameters to control the conversion between the feature vector to be output from the third neural network and the feature vector to be input to the second neural network comprises adjusting the parameters of the interface neural network, based on a relationship between the feature vector to be output from the third neural network and the feature vector to be input to the second neural network, and

the updating the parameters to control the conversion between the feature vector to be output from the first neural network and the feature vector to be input to the fourth neural network comprises adjusting the parameters of the interface neural network, based on a relationship between the feature vector to be output from the first neural network and the feature vector to be input to the fourth neural network.

6. The operation method of claim 4 , wherein the third neural network and the first neural network are distinguished with respect to any one or any combination of an input modality, an output modality, an input dimension, an output dimension, an input feature vector, and an output feature vector, and

the fourth neural network and the second neural network are distinguished with respect to any one or any combination of an input modality, an output modality, an input dimension, an output dimension, an input feature vector, and an output feature, vector.

7. The operation method of claim 4 , wherein, based on the first neural network being replaced with the third neural network, a type of an input signal based on the updated parameters is identical to a type of the input signal based on the parameters, and

based on the second neural network being replaced with the fourth neural network, a type of an output signal based on the updated parameters is identical to a type of the output signal based on the parameters.

8. The operation method of claim 1 , further comprising, based on a third neural network being additionally connected to the interface neural network, generating the parameters controlling the conversion between the feature vector to be output from the third neural network and the feature vector to be input to the second neural network.

9. The operation method of claim 1 , wherein the first neural network is configured to extract the first feature vector from an object, and

the second neural network is configured to identify the object, based on the input signal.

10. The operation method of claim 1 , wherein the first neural network is configured to determine, as the first feature vector, a command vector of an actuator, and

the second neural network is configured to control the actuator, based on the input signal.

11. A non-transitory computer-readable medium storing a program comprising instructions to control a processor to perform the method of claim 1 .

12. A signal processing apparatus comprising:

a processor configured to:

receive an output signal from a first neural network;

convert a first feature vector of the output signal to a second feature vector of an input signal to be input to a second neural network, based on parameters of an interface neural network controlling conversion between a feature vector to be output from the first neural network and a feature vector to be input to the second neural network;

generate the input signal to be input to the second neural network, the input signal comprising the second feature vector; and

transmit the input signal to the second neural network.

13. An operation method of an interface neural network, the operation method comprising:

receiving an output signal from a first neural network;

converting a first feature vector of an object and of the output signal to a second feature vector of the object and of an input signal to be input to a second neural network, based on parameters of the interface neural network that are optimized, the parameters controlling conversion between a feature vector to be output from the first neural network and a feature vector to be input to the second neural network;

generating the input signal to be input to the second neural network, the input signal comprising the second feature vector of the object;

transmitting the input signal to the second neural network; and

based on the second neural network being replaced with a third neural network, updating the parameters to control conversion between the feature vector to be output from the first neural network and a feature vector to be input to the third neural network.

14. The operation method of claim 13 , further comprising:

converting a first command of an actuator and included in the output signal to a second command of the actuator and configured to be input to the second neural network; and

generating the input signal to be input to the second neural network, the input signal comprising the second command of the actuator.