Growing Interpretable Part Graphs on ConvNets via Multi-Shot Learning

By Quanshi Zhang, Ruiming Cao, Ying Nian Wu, Song-Chun Zhu. 2016

74 citations recorded by the source corpus.

This paper proposes a learning strategy that embeds object-part concepts into a pre-trained convolutional neural network (CNN), in an attempt to 1) explore explicit semantics hidden in CNN units and 2) gradually transform the pre-trained CNN into a semantically interpretable graphical model 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 CNN units, so as to clarify their internal semantic hierarchy. Our method is guided by a small number of part annotations, and it achieves superior part-localization performance (about 13%-107% improvement in part center prediction on the PASCAL VOC and ImageNet datasets)

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Representation & Self-Supervised Learning

Citations in the corpus

Explaining Explanations: An Overview of Interpretability of Machine Learning (2018, background): “Disentangled units can enable the construction of graphs [68] and decision trees [69] to elucidate the reasoning of a network.”
Unsupervised Learning of Neural Networks to Explain Neural Networks (2018, methodology): “[41, 42] extracted certain neural units to describe an object part in a weakly-supervised manner.”
Interpreting CNNs via Decision Trees (2018, methodology): “0-2011 dataset contains 11.8K images of 200 bird species. Like in [3,21,28], we ignored species labels and regarded all these images as a single bird category. The ILSVRC 2013 DET Animal-Part dataset [28] consists of 30 animal categories among all the 200 categories for object detection in the ILSVR”
Interpretable Convolutional Neural Networks (2017, methodology): “For the ILSVRC 2013 DET Animal-Part dataset [36], we used the head and frontal legs of each category as landmarks for evaluation.”
Visual interpretability for deep learning: a survey (2018, methodology): “Thus, to provide a global view of how visual knowledge is organized in a pre-trained CNN, studies of [Zhang et al., 2018a; Zhang et al., 2016] aim to answer the following three questions.”
Examining CNN representations with respect to Dataset Bias (2017, methodology): “Methods of (Bansal, Farhadi, and Parikh 2014; Zhang et al. 2014) computed the distributions of a CNN’s prediction errors among testing samples, in order to summarize failure modes of the CNN.”
Interpreting CNN knowledge via an Explanatory Graph (2017, methodology): “Visualization of CNNs: Visualization of ﬁlters in a CNN is the most direct way of exploring the pattern hidden inside a neural unit.”
Intrinsic disentanglement: an invariance view for deep generative models (2018, background): “…filters ((Zeiler and Fergus, 2014),(Dosovitskiy and Brox, 2016)), to find correlation between intermediate feature space and data features ((Fong and Vedaldi, 2017),(Zhang et al., 2017b)) or to disentangle patterns detected by various filters to compute an explanatory graph (Zhang et al., 2017a).”
Mining Object Parts from CNNs via Active Question-Answering (2017, methodology): “AOG structure is suitable for representing semantic hierarchy of objects [50, 29], and [46] used an AOG to represent the CNN.”
From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI (2022, background): “Examples include binary maps with image patches [204], segmentation [104] and bounding boxes [297].”
Interactively Transferring CNN Patterns for Part Localization (2017, methodology): “Given the part annotations, [19, 18] mine middle-level patterns and explanatory patterns from conv-layers in the CNN to represent neural activations that are highly related to the part annotations.”
Explaining Explanations: An Approach to Evaluating Interpretability of Machine Learning (2018, background): “Disentangled units can enable the construction of graphs [68] and decision trees [69] to elucidate the reasoning of a network.”
Explainability in Deep Reinforcement Learning (2020, methodology): “They introduced the Soft Q-learning algorithm, based on maximum entropy RL [92] and energy-based models [93], as well as an extension of this algorithm that enables composition of learned skills.”
Explainable Artificial Intelligence: a Systematic Review (2020, methodology): “Lastly, And-Or Graph (AOG) [229] is a method to grow a semantic AOG on a pretrained CNN.”
Learning Decision Trees Recurrently Through Communication (2021, methodology): “To improve their performance, combining decision trees with neural networks has been explored by building hierarchical classifiers [46, 4, 65, 64], by transferring models [25, 55, 17, 23], and through regularization [61].”