| small-data-image-classification-on-cifair-10-50-samples-per-class | Cross-entropy baseline | #3 | Accuracy: 58.22 |
| small-data-image-classification-on-cifair-10-50-samples-per-class | T-vMF Similarity | #4 | Accuracy: 57.50 |
| small-data-image-classification-on-cifair-10-50-samples-per-class | Harmonic Networks | #5 | Accuracy: 56.50 |
| small-data-image-classification-on-cub-200-2011-30-samples-per-class | Harmonic Networks (no pre-training) | #2 | Accuracy: 72.26 |
| small-data-image-classification-on-cub-200-2011-30-samples-per-class | Cross-entropy baseline (no pre-training) | #3 | Accuracy: 71.44 |
| small-data-image-classification-on-cub-200-2011-30-samples-per-class | DSK Networks (no pre-training) | #4 | Accuracy: 71.02 |
| small-data-image-classification-on-deic | Harmonic Networks | #1 | Average Balanced Accuracy (across datasets): 68.70 |
| small-data-image-classification-on-deic | Cross-Entropy baseline | #2 | Average Balanced Accuracy (across datasets): 67.90 |
| small-data-image-classification-on-deic | Cosine + Cross-Entropy Loss | #3 | Average Balanced Accuracy (across datasets): 64.92 |
| small-data-image-classification-on-deic | T-vMF Similarity | #4 | Average Balanced Accuracy (across datasets): 64.67 |
| small-data-image-classification-on-deic | DSK Networks | #5 | Average Balanced Accuracy (across datasets): 64.64 |
| small-data-image-classification-on-deic | OLÉ | #6 | Average Balanced Accuracy (across datasets): 64.15 |
| small-data-image-classification-on-deic | Cosine Loss | #7 | Average Balanced Accuracy (across datasets): 62.73 |
| small-data-image-classification-on-deic | Full Convolution | #8 | Average Balanced Accuracy (across datasets): 62.06 |
| small-data-image-classification-on-deic | Deep Hybrid Networks | #9 | Average Balanced Accuracy (across datasets): 60.33 |
| small-data-image-classification-on-deic | Grad-l2 Penalty | #10 | Average Balanced Accuracy (across datasets): 55.47 |