English

Pruning for Feature-Preserving Circuits in CNNs

Computer Vision and Pattern Recognition 2023-04-18 v2 Machine Learning

Abstract

Deep convolutional neural networks are a powerful model class for a range of computer vision problems, but it is difficult to interpret the image filtering process they implement, given their sheer size. In this work, we introduce a method for extracting 'feature-preserving circuits' from deep CNNs, leveraging methods from saliency-based neural network pruning. These circuits are modular sub-functions, embedded within the network, containing only a subset of convolutional kernels relevant to a target feature. We compare the efficacy of 3 saliency-criteria for extracting these sparse circuits. Further, we show how 'sub-feature' circuits can be extracted, that preserve a feature's responses to particular images, dividing the feature into even sparser filtering processes. We also develop a tool for visualizing 'circuit diagrams', which render the entire image filtering process implemented by circuits in a parsable format.

Keywords

Cite

@article{arxiv.2206.01627,
  title  = {Pruning for Feature-Preserving Circuits in CNNs},
  author = {Chris Hamblin and Talia Konkle and George Alvarez},
  journal= {arXiv preprint arXiv:2206.01627},
  year   = {2023}
}

Comments

Under Review

R2 v1 2026-06-24T11:38:24.554Z