What Variables Affect Out-of-Distribution Generalization in Pretrained Models? Md Y ousuf Harun 1, Kyungbok Lee

Neural Information Processing Systems 

Contrary to earlier work, our experiments show this is not a universal phenomenon. We comprehensively investigate the impact of DNN architecture, training data, image resolution, and augmentations on transferability.

Similar Docs  Excel Report  more

TitleSimilaritySource
None found