SNN loss on Tensorflow MNIST example which from recently announced paper Analyzing and Improving Representations with the Soft Nearest Neighbor Loss. This is not official.See the more details in references below. If you find some error in this code, please issue to this repository
All implementations about model initialization, loss function, training, testings are in SNNloss_MNIST.ipynb
- Original paper : Frosst, Nicholas, Nicolas Papernot, and Geoffrey Hinton. "Analyzing and improving representations with the soft nearest neighbor loss." arXiv preprint arXiv:1902.01889(2019)
- Referenced source code : cleverhans - soft_nearest_neighbor_loss