Incremental Learning with Unlabeled Data in the Wild
Abstract
We propose to leverage a continuous and large stream of unlabeled data in the wild to alleviate catastrophic forget- ting in class-incremental learning. Our experimental results on CIFAR and ImageNet datasets demonstrate the superiority of the proposed methods over prior methods: compared to the state-of-the-art method, our proposed method shows up to 14.9% higher accuracy and 45.9% less forgetting.
Cite
Text
Lee et al. "Incremental Learning with Unlabeled Data in the Wild." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2019.Markdown
[Lee et al. "Incremental Learning with Unlabeled Data in the Wild." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2019.](https://mlanthology.org/cvprw/2019/lee2019cvprw-incremental/)BibTeX
@inproceedings{lee2019cvprw-incremental,
title = {{Incremental Learning with Unlabeled Data in the Wild}},
author = {Lee, Kibok and Lee, Kimin and Shin, Jinwoo and Lee, Honglak},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
year = {2019},
pages = {29-32},
url = {https://mlanthology.org/cvprw/2019/lee2019cvprw-incremental/}
}