System and method for training teacher model based on structure of student model
AssigneeSamsung Electronics Co., Ltd.
InventorsDaeyoung PARK, Changwook JEONG, Moonhyun CHA
Life of this patent
Estimating the term…
Abstract
A computer-implemented method of training a teacher model and a student model includes dividing the teacher model into a series of teacher blocks each comprising at least one layer; generating a first student branch receiving a first feature output from a first teacher block among the series of teacher blocks; training the teacher model based on outputs of the teacher model and the first student branch; and training the student model comprising a series of student blocks based on the trained teacher model, wherein the first student branch includes at least one student block.
Claims20
A training system comprising:
- a memory storing instructions;
- and
- at least one processor configured to access the memory and execute the instructions to:
- divide a teacher model into a plurality of teacher blocks each comprising at least one layer;
- generate a student branch, which receives a first feature output from a first teacher block among the plurality of teacher blocks;
- train the teacher model to decrease a difference between one or more outputs of the teacher model and an output of the student branch;
- train a student model comprising a plurality of student blocks based on the trained teacher model;
- and
- model a semiconductor process based on the trained student model.
A computer-implemented training method comprising:
- dividing a teacher model into a plurality of teacher blocks each comprising at least one layer;
- generating a student branch, which receives a first feature output from a first teacher block among the plurality of teacher blocks;
- training the teacher model to increase a similarity between one or more outputs of the teacher model and an output of the student branch;
- training a student model comprising a plurality of student blocks based on the trained teacher model;
- and
- modeling a semiconductor process based on the trained student model.
A non-transitory computer-readable storage medium comprising instructions executable by at least one processor, wherein the instructions cause the at least one processor to perform operations comprising:
- dividing a teacher model into a plurality of teacher blocks each comprising at least one layer;
- generating a student branch, which receives a first feature output from a first teacher block among the plurality of teacher blocks;
- training the teacher model to decrease a difference between one or more outputs of the teacher model and an output of the student branch;
- training a student model comprising a plurality of student blocks based on the trained teacher model;
- and
- modeling a semiconductor process based on the trained student model.
Dependencies are inferred from the claim text.