@InProceedings{BenatoTeleFalc:2021:ItPsDe,
author = "Benato, Barbara Caroline and Telea, Alexandru Cristian and
Falc{\~a}o, Alexandre Xavier",
affiliation = "{University of Campinas } and {Utrecht University } and
{University of Campinas}",
title = "Iterative Pseudo-Labeling with Deep Feature Annotation and
Confidence-Based Sampling",
booktitle = "Proceedings...",
year = "2021",
editor = "Paiva, Afonso and Menotti, David and Baranoski, Gladimir V. G. and
Proen{\c{c}}a, Hugo Pedro and Junior, Antonio Lopes Apolinario
and Papa, Jo{\~a}o Paulo and Pagliosa, Paulo and dos Santos,
Thiago Oliveira and e S{\'a}, Asla Medeiros and da Silveira,
Thiago Lopes Trugillo and Brazil, Emilio Vital and Ponti, Moacir
A. and Fernandes, Leandro A. F. and Avila, Sandra",
organization = "Conference on Graphics, Patterns and Images, 34. (SIBGRAPI)",
publisher = "IEEE Computer Society",
address = "Los Alamitos",
keywords = "semi-supervised learning, pseudolabels, optimum path forest, data
annotation.",
abstract = "Training deep neural networks is challenging when large and
annotated datasets are unavailable. Extensive manual annotation of
data samples is time-consuming, expensive, and error-prone,
notably when it needs to be done by experts. To address this
issue, increased attention has been devoted to techniques that
propagate uncertain labels (also called pseudo labels) to large
amounts of unsupervised samples and use them for training the
model. However, these techniques still need hundreds of supervised
samples per class in the training set and a validation set with
extra supervised samples to tune the model. We improve a recent
iterative pseudo-labeling technique, Deep Feature Annotation
(DeepFA), by selecting the most confident unsupervised samples to
iteratively train a deep neural network. Our confidence-based
sampling strategy relies on only dozens of annotated training
samples per class with no validation set, considerably reducing
user effort in data annotation. We first ascertain the best
configuration for the baseline a self-trained deep neural network
and then evaluate our confidence DeepFA for different confidence
thresholds. Experiments on six datasets show that DeepFA already
outperforms the self-trained baseline, but confidence DeepFA can
considerably outperform the original DeepFA and the baseline.",
conference-location = "Gramado, RS, Brazil (virtual)",
conference-year = "18-22 Oct. 2021",
doi = "10.1109/SIBGRAPI54419.2021.00034",
url = "http://dx.doi.org/10.1109/SIBGRAPI54419.2021.00034",
language = "en",
ibi = "8JMKD3MGPEW34M/45CUD68",
url = "http://urlib.net/ibi/8JMKD3MGPEW34M/45CUD68",
targetfile = "2021_sibgrapi_Benato-2.pdf",
urlaccessdate = "2024, May 06"
}