US 12,124,950 B2
Method and apparatus for optimizing quantization model, electronic device, and computer storage medium
Yi Yuan, Shenzhen (CN); Zhicheng Mao, Shenzhen (CN); Yongzhuang Wang, Shenzhen (CN); and Yuhui Xu, Shenzhen (CN)
Assigned to TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED, Shenzhen (CN)
Filed by Tencent Technology (Shenzhen) Company Limited, Shenzhen (CN)
Filed on Aug. 12, 2021, as Appl. No. 17/401,154.
Application 17/401,154 is a continuation of application No. PCT/CN2020/089543, filed on May 11, 2020.
Claims priority of application No. 201910390616.7 (CN), filed on May 10, 2019.
Prior Publication US 2021/0374540 A1, Dec. 2, 2021
Int. Cl. G06N 3/08 (2023.01); G06N 10/00 (2022.01); G06V 40/20 (2022.01)
CPC G06N 3/08 (2013.01) [G06N 10/00 (2019.01); G06V 40/20 (2022.01)] 19 Claims
OG exemplary drawing
 
1. An optimization method applicable to an electronic device, the method comprising:
determining jump ratios of embedding layer parameters of a trained quantization model in a first predetermined time range, each jump ratio corresponding to a parameter jump of the embedding layer parameters of the trained quantization model in each predetermined time interval within the first predetermined time range, the quantization model being a neural network model obtained after quantization processing on the embedding layer parameters;
determining a jump curve in the first predetermined time range according to the jump ratios;
fitting the jump curve to obtain a corresponding time scaling parameter;
optimizing an initial algorithm of the quantization model based on the time scaling parameter to obtain an optimized target optimization algorithm;
training the quantization model based on the target optimization algorithm to obtain an optimized quantization model, wherein the time scaling parameter is configured to adjust a convergence speed and a precision of the quantization model;
acquiring user behavior data associated with a user at a terminal in a second predetermined time range;
learning the user behavior data using the optimized quantization model;
determining a user behavior feature corresponding to the user behavior data;
determining target recommendation information based on the user behavior feature; and;
providing the target recommendation information to the user at the terminal.