evaluationIntroduced by True few-shot learning · 2021
True few-shot evaluation
Evaluate few-shot ability without a hidden validation set used for tuning.
Drafted by AI · not yet reviewed
How this idea evolved
Suggested from citations. No curator has recorded what this idea builds on yet. These ideas from the same theme come from papers that True few-shot learning cites, directly or one step removed. Citation is a fact; the connection between the ideas is not verified.
- ImageNet accuracy predicts transferDo better ImageNet models transfer bette · 2018two citations away
Better ImageNet models are, almost linearly, better feature extractors elsewhere.
Accuracy drops on new test sets drawn the same way; robustness to such shifts is rare.
ImageNet CNNs recognise textures more than shapes, unlike humans.
Papers using this
No other paper in this dataset is tagged with it yet.