# Subtractor-Based CNN Inference Accelerator

arXiv: https://arxiv.org/abs/2310.01022 · HTML: https://mlsanity.com/paper/2310.01022/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 32 references that are in the map.
- First posted on arXiv 2 Oct 2023.

## Builds on (most cited first, 7 of 7)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference](https://mlsanity.com/paper/1712.05877/) (2017)
- [ResNet Can Be Pruned 60x: Introducing Network Purification and Unused Path Removal (P-RM) after Weight Pruning](https://mlsanity.com/paper/1905.00136/) (2019)
- [Deep Learning Training with Simulated Approximate Multipliers](https://mlsanity.com/paper/2001.00060/) (2019)
- [Using Deep Learning to Automate the Detection of Flaws in Nuclear Fuel Channel UT Scans](https://mlsanity.com/paper/2102.13635/) (2021)
- [A Comparative Study on Machine Learning Algorithms for the Control of a Wall Following Robot](https://mlsanity.com/paper/1912.11856/) (2019)

## Abstract (arXiv)

This paper presents a novel method to boost the performance of CNN inference accelerators by utilizing subtractors. The proposed CNN preprocessing accelerator relies on sorting, grouping, and rounding the weights to create combinations that allow for the replacement of one multiplication operation and addition operation by a single subtraction operation when applying convolution during inference. Given the high cost of multiplication in terms of power and area, replacing it with subtraction allows for a performance boost by reducing power and area. The proposed method allows for controlling the trade-off between performance gains and accuracy loss through increasing or decreasing the usage of subtractors. With a rounding size of 0.05 and by utilizing LeNet-5 with the MNIST dataset, the proposed design can achieve 32.03% power savings and a 24.59% reduction in area at the cost of only 0.1% in terms of accuracy loss.