Open AccessUsing instance hardness measures in curriculum learning
Author(s) -
Gustavo Henrique Nunes,
Gustavo O. Martins,
Carlos Henrique Quartucci Forster,
Ana Carolina Lorena
Publication year - 2021
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5753/eniac.2021.18251
Subject(s) - computer science , curriculum , artificial intelligence , machine learning , convergence (economics) , order (exchange) , psychology , pedagogy , finance , economics , economic growth
Curriculum learning consists of training strategies for machine learning techniques in which the easiest observations are presented first, progressing into more dicult cases as training proceeds. For assembling the curriculum, it is necessary to order the observations a dataset has according to their diculty. This work investigates how instance hardness measures, which can be used to assess the diculty level of each observation in a dataset from dierent perspectives, can be used to assemble a curriculum. Experiments with four CIFAR-100 sub-problems have demonstrated the feasibility of using the instance hardness measures, the main advantage is on convergence speed and some datasets accuracy gains can also be verified.