METHOD OF CONTINUAL-LEARNING OF DATA SETS AND APPARATUS THEREOF

摘要

Disclosed are a method and an apparatus for sequentially performing machine learning of data sets. The method comprises the steps of: generating a first feature space based on a first data set, generating first predicted label information based on the first feature space, performing machine learning on the relationship between the first data set and the first label information associated therewith, and performing machine learning on the relationship between the first predicted label information and the first feature space to generate a first machine learning model; and generating a second feature space based on the first machine learning model and a second data set, generating second predicted label information from the second feature space, performing machine learning on the relationship between the second data set and the second label information, and performing machine learning on the relationship between the second predicted label information and the second feature space to generate a second machine learning model.