# NNK-Means: Data summarization using dictionary learning with non-negative kernel regression

arXiv: https://arxiv.org/abs/2110.08212 · HTML: https://mlsanity.com/paper/2110.08212/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 35 references that are in the map.
- First posted on arXiv 15 Oct 2021.

## Builds on (most cited first, 5 of 5)

- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Fair k-Center Clustering for Data Summarization](https://mlsanity.com/paper/1901.08628/) (2019)
- [Coresets and Sketches](https://mlsanity.com/paper/1601.00617/) (2016)
- [Fast Low-rank Shared Dictionary Learning for Image Classification](https://mlsanity.com/paper/1610.08606/) (2016)
- [Linearized Kernel Dictionary Learning](https://mlsanity.com/paper/1509.05634/) (2015)

## Built on by (most cited first, 2 of 2)

- [Out-of-Distribution Detection through Soft Clustering with Non-Negative Kernel Regression](https://mlsanity.com/paper/2407.13141/) (2024)
- [Towards Automated Error Discovery: A Study in Conversational AI](https://mlsanity.com/paper/2509.10833/) (2025)

## Abstract (arXiv)

An increasing number of systems are being designed by gathering significant amounts of data and then optimizing the system parameters directly using the obtained data. Often this is done without analyzing the dataset structure. As task complexity, data size, and parameters all increase to millions or even billions, data summarization is becoming a major challenge. In this work, we investigate data summarization via dictionary learning~(DL), leveraging the properties of recently introduced non-negative kernel regression (NNK) graphs. Our proposed NNK-Means, unlike previous DL techniques, such as kSVD, learns geometric dictionaries with atoms that are representative of the input data space. Experiments show that summarization using NNK-Means can provide better class separation compared to linear and kernel versions of kMeans and kSVD. Moreover, NNK-Means is scalable, with runtime complexity similar to that of kMeans.