Example Perplexity
This work addresses the challenge of understanding example-specific classification difficulty for researchers and practitioners in machine learning, but it appears incremental as it builds on existing concepts of difficulty without claiming major breakthroughs.
The paper tackles the problem of measuring classification difficulty for individual examples in deep neural networks, introducing 'example perplexity' as a metric and analyzing factors that contribute to high perplexity, with no concrete numbers provided in the abstract.
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.