RotNet

Reported on 2 benchmarks across 2 tasks · 2 papers

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

Computer Vision1 result

Image ClassificationonSTL-10
Percentage correct· 2020-06-22
68.19
best: 99.64 (µ2Net+ (ViT-L/16))
Don't Wait, Just Weight: Improving Unsupervised Representations by Learning Goal-Driven Instance Weights arXiv:2006.12360

Methodology1 result

Anomaly DetectiononOne-class ImageNet-30
AUROC· 2019-06-28
65.3
best: 99.9 (BCE-Clip (OE))
Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty arXiv:1906.12340