SCARLET-B

Reported on 5 benchmarks across 3 tasks · 1 paper · 2 SOTA

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

Methodology4 results

Neural Architecture SearchonImageNet
Top-1 Error Rate· 2019-08-16
23.7
best: 16.1 (DeepMAD-50M)
SOTA
SCARLET-NAS: Bridging the Gap between Stability and Scalability in Weight-sharing Neural Architecture Search arXiv:1908.06022
AutoMLonImageNet
Top-1 Error Rate· 2019-08-16
23.7
best: 16.1 (DeepMAD-50M)
SOTA
SCARLET-NAS: Bridging the Gap between Stability and Scalability in Weight-sharing Neural Architecture Search arXiv:1908.06022
Neural Architecture SearchonImageNet
Accuracy· 2019-08-16
76.3
best: 83.9 (DeepMAD-50M)
SCARLET-NAS: Bridging the Gap between Stability and Scalability in Weight-sharing Neural Architecture Search arXiv:1908.06022
AutoMLonImageNet
Accuracy· 2019-08-16
76.3
best: 83.9 (DeepMAD-50M)
SCARLET-NAS: Bridging the Gap between Stability and Scalability in Weight-sharing Neural Architecture Search arXiv:1908.06022

Computer Vision1 result

Image ClassificationonImageNet
GFLOPs· 2019-08-16
0.658
best: 1478 (InternImage-H)
SCARLET-NAS: Bridging the Gap between Stability and Scalability in Weight-sharing Neural Architecture Search arXiv:1908.06022