Some examples are easier for humans to classify than others. The same should be true for deep neural networks (DNNs). We use the term example perplexity to refer to the level of difficulty of classifying an example. In this paper, we propose a method to measure the perplexity of an example and investigate what factors contribute to high example perplexity. The related codes and resources are available at https://github.com/vaynexie/Example-Perplexity.
@article{arxiv.2203.08813,
title = {Example Perplexity},
author = {Nevin L. Zhang and Weiyan Xie and Zhi Lin and Guanfang Dong and Xiao-Hui Li and Caleb Chen Cao and Yunpeng Wang},
journal= {arXiv preprint arXiv:2203.08813},
year = {2022}
}