# Understanding the Impact of Label Granularity on CNN-based Image Classification

arXiv: https://arxiv.org/abs/1901.07012 · HTML: https://mlsanity.com/paper/1901.07012/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Jan 2019.

## Built on by (most cited first, 13 of 13)

- [Efficient Algorithms for Learning from Coarse Labels](https://mlsanity.com/paper/2108.09805/) (2021)
- [Subtype-aware Unsupervised Domain Adaptation for Medical Diagnosis](https://mlsanity.com/paper/2101.00318/) (2021)
- [Self-Supervised Vision Transformers for Malware Detection](https://mlsanity.com/paper/2208.07049/) (2022)
- [On the Efficiency of Subclass Knowledge Distillation in Classification Tasks](https://mlsanity.com/paper/2109.05587/) (2021)
- [Towards Understanding the Effect of Pretraining Label Granularity](https://mlsanity.com/paper/2303.16887/) (2023)
- [Subtype-Aware Dynamic Unsupervised Domain Adaptation](https://mlsanity.com/paper/2208.07754/) (2022)
- [Loss Functions for Classification using Structured Entropy](https://mlsanity.com/paper/2206.07122/) (2022)
- [BC-MRI-SEG: A Breast Cancer MRI Tumor Segmentation Benchmark](https://mlsanity.com/paper/2404.13756/) (2024)
- [The Advantage of Fine-Grained Training](https://mlsanity.com/paper/2509.05130/) (2025)
- [Extremely coarse learning objectives induce human-aligned representations in AI vision models](https://mlsanity.com/paper/2605.05556/) (2026)
- [Deep Learning based Prediction of MSI using MMR Markers in Colorectal Cancer](https://mlsanity.com/paper/2203.00449/) (2022)
- [Generating Realistic, Diverse, and Fault-Revealing Inputs with Latent Space Interpolation for Testing Deep Neural Networks](https://mlsanity.com/paper/2503.17630/) (2025)
- [When Tokenization is Secretly Output Supervision](https://mlsanity.com/paper/2609.01386/) (2026)

## Abstract (arXiv)

In recent years, supervised learning using Convolutional Neural Networks (CNNs) has achieved great success in image classification tasks, and large scale labeled datasets have contributed significantly to this achievement. However, the definition of a label is often application dependent. For example, an image of a cat can be labeled as "cat" or perhaps more specifically "Persian cat." We refer to this as label granularity. In this paper, we conduct extensive experiments using various datasets to demonstrate and analyze how and why training based on fine-grain labeling, such as "Persian cat" can improve CNN accuracy on classifying coarse-grain classes, in this case "cat." The experimental results show that training CNNs with fine-grain labels improves both network's optimization and generalization capabilities, as intuitively it encourages the network to learn more features, and hence increases classification accuracy on coarse-grain classes under all datasets considered. Moreover, fine-grain labels enhance data efficiency in CNN training. For example, a CNN trained with fine-grain labels and only 40% of the total training data can achieve higher accuracy than a CNN trained with the full training dataset and coarse-grain labels. These results point to two possible applications of this work: (i) with sufficient human resources, one can improve CNN performance by re-labeling the dataset with fine-grain labels, and (ii) with limited human resources, to improve CNN performance, rather than collecting more training data, one may instead use fine-grain labels for the dataset. We further propose a metric called Average Confusion Ratio to characterize the effectiveness of fine-grain labeling, and show its use through extensive experimentation. Code is available at https://github.com/cmu-enyac/Label-Granularity.