Balanced Softmax (BALMS)

Reported on 5 benchmarks across 5 tasks · 1 paper · 5 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology3 results

Generalized Few-Shot ClassificationonCIFAR-10-LT (ρ=10)
Error Rate· 2020-07-21
8.7
best: 5 (GLMC+MaxNorm (ResNet-34, channel x4))
SOTA
Balanced Meta-Softmax for Long-Tailed Visual Recognition arXiv:2007.10740
Long-tail LearningonCIFAR-10-LT (ρ=10)
Error Rate· 2020-07-21
8.7
best: 5 (GLMC+MaxNorm (ResNet-34, channel x4))
SOTA
Balanced Meta-Softmax for Long-Tailed Visual Recognition arXiv:2007.10740
Generalized Few-Shot LearningonCIFAR-10-LT (ρ=10)
Error Rate· 2020-07-21
8.7
best: 5 (GLMC+MaxNorm (ResNet-34, channel x4))
SOTA
Balanced Meta-Softmax for Long-Tailed Visual Recognition arXiv:2007.10740

Computer Vision2 results

Image ClassificationonCIFAR-10-LT (ρ=10)
Error Rate· 2020-07-21
8.7
best: 5 (GLMC+MaxNorm (ResNet-34, channel x4))
SOTA
Balanced Meta-Softmax for Long-Tailed Visual Recognition arXiv:2007.10740
Few-Shot Image ClassificationonCIFAR-10-LT (ρ=10)
Error Rate· 2020-07-21
8.7
best: 5 (GLMC+MaxNorm (ResNet-34, channel x4))
SOTA
Balanced Meta-Softmax for Long-Tailed Visual Recognition arXiv:2007.10740