ProgressiveNet

Reported on 6 benchmarks across 1 task · 1 paper · 6 SOTA

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

Methodology6 results

Continual LearningonSketch (Fine-grained 6 Tasks)
Accuracy· 2016-06-15
76.35
best: 80.77 (CondConvContinual)
SOTA
Progressive Neural Networks arXiv:1606.04671
Continual LearningonStanford Cars (Fine-grained 6 Tasks)
Accuracy· 2016-06-15
89.21
best: 92.8 (CPG)
SOTA
Progressive Neural Networks arXiv:1606.04671
Continual LearningonCUBS (Fine-grained 6 Tasks)
Accuracy· 2016-06-15
78.94
best: 84.26 (CondConvContinual)
SOTA
Progressive Neural Networks arXiv:1606.04671
Continual LearningonWikiart (Fine-grained 6 Tasks)
Accuracy· 2016-06-15
74.94
best: 78.32 (CondConvContinual)
SOTA
Progressive Neural Networks arXiv:1606.04671
Continual LearningonImageNet (Fine-grained 6 Tasks)
Accuracy· 2016-06-15
76.16
SOTA
Progressive Neural Networks arXiv:1606.04671
Continual LearningonFlowers (Fine-grained 6 Tasks)
Accuracy· 2016-06-15
93.41
best: 97.16 (CondConvContinual)
SOTA
Progressive Neural Networks arXiv:1606.04671