# MACE: Model Agnostic Concept Extractor for Explaining Image Classification Networks

arXiv: https://arxiv.org/abs/2011.01472 · HTML: https://mlsanity.com/paper/2011.01472/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Nov 2020.

## Built on by (most cited first, 6 of 6)

- [Multi-dimensional concept discovery (MCD): A unifying framework with completeness guarantees](https://mlsanity.com/paper/2301.11911/) (2023)
- [Sparse Subspace Clustering for Concept Discovery (SSCCD)](https://mlsanity.com/paper/2203.06043/) (2022)
- [Scalable Concept Extraction in Industry 4.0](https://mlsanity.com/paper/2306.03551/) (2023)
- [GIFT: A Framework Towards Global Interpretable Faithful Textual Explanations of Vision Classifiers](https://mlsanity.com/paper/2411.15605/) (2024)
- [Explainable Supervised Domain Adaptation](https://mlsanity.com/paper/2205.09943/) (2022)
- [P-TAME: Explain Any Image Classifier with Trained Perturbations](https://mlsanity.com/paper/2501.17813/) (2025)

## Abstract (arXiv)

Deep convolutional networks have been quite successful at various image classification tasks. The current methods to explain the predictions of a pre-trained model rely on gradient information, often resulting in saliency maps that focus on the foreground object as a whole. However, humans typically reason by dissecting an image and pointing out the presence of smaller concepts. The final output is often an aggregation of the presence or absence of these smaller concepts. In this work, we propose MACE: a Model Agnostic Concept Extractor, which can explain the working of a convolutional network through smaller concepts. The MACE framework dissects the feature maps generated by a convolution network for an image to extract concept based prototypical explanations. Further, it estimates the relevance of the extracted concepts to the pre-trained model's predictions, a critical aspect required for explaining the individual class predictions, missing in existing approaches. We validate our framework using VGG16 and ResNet50 CNN architectures, and on datasets like Animals With Attributes 2 (AWA2) and Places365. Our experiments demonstrate that the concepts extracted by the MACE framework increase the human interpretability of the explanations, and are faithful to the underlying pre-trained black-box model.