Abstract: Data augmentation is an effective technique to enrich the training data’s diversity and reduce the risk of overfitting. However, different data sets have distinct preferences on various ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results