@inproceedings{barz2020deep,
    type = {inproceedings},
    key = {barz2020deep},
    title = {Deep Learning on Small Datasets without Pre-Training using Cosine Loss},
    author = {Björn Barz and Joachim Denzler},
    booktitle = {IEEE Winter Conference on Applications of Computer Vision (WACV)},
    year = {2020},
    pages = {1360-1369},
    abstract = {Two things seem to be indisputable in the contemporary deep learning discourse: 1. The categorical cross-entropy loss after softmax activation is the method of choice for classification. 2. Training a CNN classifier from scratch on small datasets does not work well. In contrast to this, we show that the cosine loss function provides substantially better performance than cross-entropy on datasets with only a handful of samples per class. For example, the accuracy achieved on the CUB-200-2011 dataset without pre-training is by 30% higher than with the cross-entropy loss. Further experiments on other popular datasets confirm our findings. Moreover, we demonstrate that integrating prior knowledge in the form of class hierarchies is straightforward with the cosine loss and improves classification performance further.},
    code = {https://github.com/cvjena/semantic-embeddings/blob/v1.2.0/CosineLoss.md},
    doi = {10.1109/WACV45572.2020.9093286},
    groups = {deeplearning},
}