# Growing Interpretable Part Graphs on ConvNets via Multi-Shot Learning

arXiv: https://arxiv.org/abs/1611.04246 · HTML: https://mlsanity.com/paper/1611.04246/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Nov 2016.

## Built on by (most cited first, 10 of 10)

- [From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI](https://mlsanity.com/paper/2201.08164/) (2022)
- [Using Logic Programming and Kernel-Grouping for Improving Interpretability of Convolutional Neural Networks](https://mlsanity.com/paper/2310.13073/) (2023)
- [Finding Representative Interpretations on Convolutional Neural Networks](https://mlsanity.com/paper/2108.06384/) (2021)
- [NeurFlow: Interpreting Neural Networks through Neuron Groups and Functional Interactions](https://mlsanity.com/paper/2502.16105/) (2025)
- [NeSyFOLD: Neurosymbolic Framework for Interpretable Image Classification](https://mlsanity.com/paper/2301.12667/) (2023)
- [A Survey on Interpretability in Visual Recognition](https://mlsanity.com/paper/2507.11099/) (2025)
- [Enhancing Pre-trained Representation Classifiability can Boost its Interpretability](https://mlsanity.com/paper/2510.24105/) (2025)
- [Explainability-aided Domain Generalization for Image Classification](https://mlsanity.com/paper/2104.01742/) (2021)
- [AutoProtoNet: Interpretability for Prototypical Networks](https://mlsanity.com/paper/2204.00929/) (2022)
- [A Neurosymbolic Framework for Bias Correction in Convolutional Neural Networks](https://mlsanity.com/paper/2405.15886/) (2024)

## Abstract (arXiv)

This paper proposes a learning strategy that extracts object-part concepts from a pre-trained convolutional neural network (CNN), in an attempt to 1) explore explicit semantics hidden in CNN units and 2) gradually grow a semantically interpretable graphical model on the pre-trained CNN for hierarchical object understanding. Given part annotations on very few (e.g., 3-12) objects, our method mines certain latent patterns from the pre-trained CNN and associates them with different semantic parts. We use a four-layer And-Or graph to organize the mined latent patterns, so as to clarify their internal semantic hierarchy. Our method is guided by a small number of part annotations, and it achieves superior performance (about 13%-107% improvement) in part center prediction on the PASCAL VOC and ImageNet datasets.