Curriculum learning is a bio-inspired training technique that is widely adopted to machine learning for improved optimization and better training of neural networks regarding the convergence rate or obtained accuracy. The main concept in curriculum learning is to start the training with simpler tasks and gradually increase the level of difficulty. Therefore, a natural question is how to determine or generate these simpler tasks. In this work, we take inspiration from Spatial Transformer Networks (STNs) in order to form an easy-to-hard curriculum. As STNs have been proven to be capable of removing the clutter from the input images and obtaining higher accuracy in image classification tasks, we hypothesize that images processed by STNs can be seen as easier tasks and utilized in the interest of curriculum learning. To this end, we study multiple strategies developed for shaping the training curriculum, using the data generated by STNs. We perform various experiments on cluttered MNIST and Fashion-MNIST datasets, where on the former, we obtain an improvement of $3.8$pp in classification accuracy compared to the baseline.
@article{pub11864,
author = {
Azimi, Fatemeh
and
Nies, Jean-Francois
and
Palacio, Sebastian
and
Raue, Federico
and
Hees, Jörn
and
Dengel, Andreas
},
title = {Spatial Transformer Networks for Curriculum Learning},
booktitle = {ArXiv},
year = {2021},
volume = {abs/2108.09696},
journal = {Computing Research Repository eprint Journal (CoRR)},
publisher = {ArXiv}
}
Deutsches Forschungszentrum für Künstliche Intelligenz German Research Center for Artificial Intelligence