# Energy Efficient Hardware Acceleration of Neural Networks with Power-of-Two Quantisation

arXiv: https://arxiv.org/abs/2209.15257 · HTML: https://mlsanity.com/paper/2209.15257/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 10 references that are in the map.
- First posted on arXiv 30 Sep 2022.

## Builds on (most cited first, 6 of 6)

- [A Survey of Quantization Methods for Efficient Neural Network Inference](https://mlsanity.com/paper/2103.13630/) (2021)
- [Additive Powers-of-Two Quantization: An Efficient Non-uniform Discretization for Neural Networks](https://mlsanity.com/paper/1909.13144/) (2019)
- [Convolutional Neural Networks using Logarithmic Data Representation](https://mlsanity.com/paper/1603.01025/) (2016)
- [DeepShift: Towards Multiplication-Less Neural Networks](https://mlsanity.com/paper/1905.13298/) (2019)
- [Accurate Neural Training with 4-bit Matrix Multiplications at Standard Formats](https://mlsanity.com/paper/2112.10769/) (2021)
- [Power-of-Two Quantization for Low Bitwidth and Hardware Compliant Neural Networks](https://mlsanity.com/paper/2203.05025/) (2022)

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

- [PowerYOLO: Mixed Precision Model for Hardware Efficient Object Detection with Event Data](https://mlsanity.com/paper/2407.08272/) (2024)
- [Probabilistic Weight Fixing: Large-scale training of neural network weight uncertainties for quantization](https://mlsanity.com/paper/2309.13575/) (2023)
- [Accelerating PoT Quantization on Edge Devices](https://mlsanity.com/paper/2409.20403/) (2024)
- [PoTAcc: A Pipeline for End-to-End Acceleration of Power-of-Two Quantized DNNs](https://mlsanity.com/paper/2605.06082/) (2026)

## Abstract (arXiv)

Deep neural networks virtually dominate the domain of most modern vision systems, providing high performance at a cost of increased computational complexity.Since for those systems it is often required to operate both in real-time and with minimal energy consumption (e.g., for wearable devices or autonomous vehicles, edge Internet of Things (IoT), sensor networks), various network optimisation techniques are used, e.g., quantisation, pruning, or dedicated lightweight architectures. Due to the logarithmic distribution of weights in neural network layers, a method providing high performance with significant reduction in computational precision (for 4-bit weights and less) is the Power-of-Two (PoT) quantisation (and therefore also with a logarithmic distribution). This method introduces additional possibilities of replacing the typical for neural networks Multiply and ACcumulate (MAC -- performing, e.g., convolution operations) units, with more energy-efficient Bitshift and ACcumulate (BAC). In this paper, we show that a hardware neural network accelerator with PoT weights implemented on the Zynq UltraScale + MPSoC ZCU104 SoC FPGA can be at least 1.4x more energy efficient than the uniform quantisation version. To further reduce the actual power requirement by omitting part of the computation for zero weights, we also propose a new pruning method adapted to logarithmic quantisation.