SELF-SUPERVISED LEARNING OF A TASK WITH NORMALIZATION OF NUISANCE FROM A DIFFERENT TASK

A computer-implemented method including: obtaining a pre-trained upstream machine learning model; fine-tuning the pre-trained upstream model for at least two downstream tasks, wherein the fine-tuning comprises: (a) training a target downstream model for a target downstream task, based on a dataset w...

Full description

Saved in:
Bibliographic Details
Main Authors Aronowitz, Hagai, Gat, Itai
Format Patent
LanguageEnglish
Published 06.06.2024
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:A computer-implemented method including: obtaining a pre-trained upstream machine learning model; fine-tuning the pre-trained upstream model for at least two downstream tasks, wherein the fine-tuning comprises: (a) training a target downstream model for a target downstream task, based on a dataset with labeling specific to the target downstream task, and (b) training a nuisance downstream model for a nuisance downstream task, based on a dataset with labeling of characteristics which are specific to the nuisance downstream task and are undesired for the target downstream task; and normalizing the undesired characteristics from the pre-trained upstream model, to prevent biasing of the target downstream model by the undesired characteristics.
Bibliography:Application Number: US202218074536