Some tasks are inferred based on the benchmarks list.
The benchmarks section lists all benchmarks using a given dataset or any of its variants. We use variants to distinguish between results evaluated on slightly different versions of the same dataset. For example, ImageNet 32⨉32 and ImageNet 64⨉64 are variants of the ImageNet dataset.
The following are all the runs used to generate figures in the paper. Every experiment solves the corresponding high- and low-fidelity model to generate the training, validation, and prediction data.