Year: 2022
Type: preprint
Abstract: It was shown that adversarial examples improve object recognition. But what about their opposite side, easy examples? Easy examples are samples that the machine learning model classifies correctly with high confidence. In our paper, we are making the first step toward exploring the potential benefits of using easy examples in the training procedure of neural networks. We propose to use an auxiliary batch normalization for easy examples for the standard and robust accuracy improvement.
Source: arXiv (Cornell University)
Authors: Asadulaev, Arip, Panfilov, Alexander, Filchenkov, Andrey
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4286231741