Method and apparatus with neural network parameter quantization
AssigneesSamsung Electronics Co., Ltd., Unist (Ulsan National Institute of Science and Technology)
InventorsHyeongseok Yu, Hyeonuk Sim, Jongeun Lee
Life of this patent
Estimating the term…
Abstract
A processor-implemented method includes determining a first quantization value by performing log quantization on a parameter from one of input activation values and weight values in a layer of a neural network, comparing a threshold value with an error between a first dequantization value obtained by dequantization of the first quantization value and the parameter, determining a second quantization value by performing log quantization on the error in response to the error being greater than the threshold value as a result of the comparing; and quantizing the parameter to a value in which the first quantization value and the second quantization value are grouped.
Claims19
A processor-implemented method, the method comprising:
- determining a first quantization value by performing log quantization on a parameter from one of input activation values and weight values in a layer of a neural network;
- comparing a threshold value with an error between a first dequantization value obtained by dequantization of the first quantization value and the parameter;
- determining a second quantization value by performing log quantization on the error in response to the error being greater than the threshold value as a result of the comparing;
- and
- quantizing the parameter to a value in which the first quantization value and the second quantization value are grouped.
An apparatus, the apparatus comprising:
- one or more processors configured to:
- determine a first quantization value by performing log quantization on a parameter from one of input activation values and weight values in a layer of a neural network;
- compare a threshold value with an error between a first dequantization value obtained by dequantization of the first quantization value and the parameter;
- determine a second quantization value by performing log quantization on the error in response to the error being greater than the threshold value as a result of the comparing;
- and
- quantize the parameter to a value in which the first quantization value and the second quantization value are grouped.
Dependencies are inferred from the claim text.