METHOD AND APPARATUS FOR IMPROVING EFFECTIVE PRECISION OF NEURAL NETWORK THROUGH ARCHITECTURE EXTENSION

1 - 10 . (canceled)

11 . A computer-implemented method for expanding an architecture of a neural network, the method comprising:

selecting a target producer neuron from neurons included in the neural network, the target producer neuron outputting an activation clipped according to a given clipping range;

dividing the given clipping range into a plurality of segments;

replacing the target producer neuron with a plurality of producer neurons corresponding to the segments;

setting parameters of each producer neuron such that each producer neuron processes an input of the target producer neuron; and

setting parameters of a consumer neuron connected to the target producer neuron such that the consumer neuron processes outputs of the plurality of producer neurons.

12 . The method according to claim 11 , wherein each producer neuron outputs an activation clipped according to a range of a corresponding segment among the plurality of segments.

13 . The method according to claim 11 , wherein a plurality of output activations output by the plurality of producer neurons has the same precision as precision of the clipped activation output by the target producer neuron.

14 . The method according to claim 11 , further comprising converting the plurality of segments into segments that have the same size as the given clipping range and do not overlap.

15 . The method according to claim 11 , wherein at least two of the plurality of segments have different sizes.

16 . The method according to claim 11 , further comprising adjusting the range of each segment in consideration of a computation range of each producer neuron.

17 . The method according to claim 16 , wherein the setting of parameters of each producer neuron comprises setting parameters of each producer neuron based on the range of the segment corresponding to each producer neuron and the adjusted range of the segment, and

the setting of parameters of the consumer neuron comprises setting parameters applied to the output of each producer neuron based on the range of the segment corresponding to each producer neuron and the adjusted range of the segment.

18 . The method according to claim 11 , wherein each producer neuron processes the input using the same parameters as parameters of the target producer neuron, and the consumer neuron processes the outputs of the plurality of producer neurons using the same parameters as parameters applied to the output of the target producer and an offset according to the plurality of segments.

19 . A computing device comprising:

a memory in which instructions are stored; and

at least one processor,

wherein the at least one processor is configured to, by executing the instructions:

select a target producer neuron from the neurons included in a neural network, the target producer neuron outputting an activation clipped according to a given clipping range;

divide the given clipping range into a plurality of segments;

replace the target producer neuron with a plurality of producer neurons corresponding to the segments;

set parameters of each producer neuron such that each producer neuron processes an input of the target producer neuron; and

set parameters of a consumer neuron connected to the target producer neuron such that the consumer neuron processes outputs of the plurality of producer neurons.

20 . A computer-readable recording medium recording a computer program for executing the method of claim 11 .