# Spectral Metric for Dataset Complexity Assessment

arXiv: https://arxiv.org/abs/1905.07299 · HTML: https://mlsanity.com/paper/1905.07299/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 May 2019.

## Built on by (most cited first, 15 of 15)

- [Data Quality Measures and Efficient Evaluation Algorithms for Large-Scale High-Dimensional Data](https://mlsanity.com/paper/2101.01441/) (2021)
- [Adversarial Robustness in Deep Learning: Attacks on Fragile Neurons](https://mlsanity.com/paper/2201.12347/) (2022)
- [problexity -- an open-source Python library for binary classification problem complexity assessment](https://mlsanity.com/paper/2207.06709/) (2022)
- [MOTCOM: The Multi-Object Tracking Dataset Complexity Metric](https://mlsanity.com/paper/2207.10031/) (2022)
- [Complexity of Representations in Deep Learning](https://mlsanity.com/paper/2209.00525/) (2022)
- [PINCH: An Adversarial Extraction Attack Framework for Deep Learning Models](https://mlsanity.com/paper/2209.06300/) (2022)
- [Fragility, Robustness and Antifragility in Deep Learning](https://mlsanity.com/paper/2312.09821/) (2023)
- [Adaptive Neighbourhoods for the Discovery of Adversarial Examples](https://mlsanity.com/paper/2101.09108/) (2021)
- [ISTRBoost: Importance Sampling Transfer Regression using Boosting](https://mlsanity.com/paper/2204.12044/) (2022)
- [Dataset Complexity Assessment Based on Cumulative Maximum Scaled Area Under Laplacian Spectrum](https://mlsanity.com/paper/2209.14743/) (2022)
- [Poison is Not Traceless: Fully-Agnostic Detection of Poisoning Attacks](https://mlsanity.com/paper/2310.16224/) (2023)
- [Analytical Softmax Temperature Setting from Feature Dimensions for Model- and Domain-Robust Classification](https://mlsanity.com/paper/2504.15594/) (2025)
- [Evaluating Knowledge Graph Complexity via Semantic, Spectral, and Structural Metrics for Link Prediction](https://mlsanity.com/paper/2508.15291/) (2025)
- [Evaluating Cumulative Spectral Gradient as a Complexity Measure](https://mlsanity.com/paper/2509.02399/) (2025)
- [Meta-Learning for Classifier Selection in Image Datasets: A Feature-Driven Framework for Accuracy Prediction](https://mlsanity.com/paper/2609.11041/) (2026)

## Abstract (arXiv)

In this paper, we propose a new measure to gauge the complexity of image classification problems. Given an annotated image dataset, our method computes a complexity measure called the cumulative spectral gradient (CSG) which strongly correlates with the test accuracy of convolutional neural networks (CNN). The CSG measure is derived from the probabilistic divergence between classes in a spectral clustering framework. We show that this metric correlates with the overall separability of the dataset and thus its inherent complexity. As will be shown, our metric can be used for dataset reduction, to assess which classes are more difficult to disentangle, and approximate the accuracy one could expect to get with a CNN. Results obtained on 11 datasets and three CNN models reveal that our method is more accurate and faster than previous complexity measures.