# Spatial Correlation and Value Prediction in Convolutional Neural Networks

arXiv: https://arxiv.org/abs/1807.10598 · HTML: https://mlsanity.com/paper/1807.10598/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Jul 2018.

## Built on by (most cited first, 4 of 4)

- [WIR3D: Visually-Informed and Geometry-Aware 3D Shape Abstraction](https://mlsanity.com/paper/2505.04813/) (2025)
- [Intelligent Icing Detection Model of Wind Turbine Blades Based on SCADA data](https://mlsanity.com/paper/2101.07914/) (2021)
- [Mixture-of-Rookies: Saving DNN Computations by Predicting ReLU Outputs](https://mlsanity.com/paper/2202.04990/) (2022)
- [Layer-wise Input Recoverability in Deep Neural Networks](https://mlsanity.com/paper/2610.14290/) (2026)

## Abstract (arXiv)

Convolutional neural networks (CNNs) are a widely used form of deep neural networks, introducing state-of-the-art results for different problems such as image classification, computer vision tasks, and speech recognition. However, CNNs are compute intensive, requiring billions of multiply-accumulate (MAC) operations per input. To reduce the number of MACs in CNNs, we propose a value prediction method that exploits the spatial correlation of zero-valued activations within the CNN output feature maps, thereby saving convolution operations. Our method reduces the number of MAC operations by 30.4%, averaged on three modern CNNs for ImageNet, with top-1 accuracy degradation of 1.7%, and top-5 accuracy degradation of 1.1%.