# PositNN: Training Deep Neural Networks with Mixed Low-Precision Posit

arXiv: https://arxiv.org/abs/2105.00053 · HTML: https://mlsanity.com/paper/2105.00053/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 23 references that are in the map.
- First posted on arXiv 30 Apr 2021.

## Builds on (most cited first, 10 of 10)

- [Deep Learning in Neural Networks: An Overview](https://mlsanity.com/paper/1404.7828/) (2014)
- [The Computational Limits of Deep Learning](https://mlsanity.com/paper/2007.05558/) (2020)
- [Rethinking floating point for deep learning](https://mlsanity.com/paper/1811.01721/) (2018)
- [Deep Positron: A Deep Neural Network Using the Posit Number System](https://mlsanity.com/paper/1812.01762/) (2018)
- [Cheetah: Mixed Low-Precision Hardware & Software Co-Design Framework for DNNs on the Edge](https://mlsanity.com/paper/1908.02386/) (2019)
- [Performance-Efficiency Trade-off of Low-Precision Numerical Formats in Deep Neural Networks](https://mlsanity.com/paper/1903.10584/) (2019)
- [Training Deep Neural Networks Using Posit Number System](https://mlsanity.com/paper/1909.03831/) (2019)
- [Deep Learning Inference on Embedded Devices: Fixed-Point vs Posit](https://mlsanity.com/paper/1805.08624/) (2018)
- [Deep Learning Training on the Edge with Low-Precision Posits](https://mlsanity.com/paper/1907.13216/) (2019)
- [Template-Based Posit Multiplication for Training and Inferring in Neural Networks](https://mlsanity.com/paper/1907.04091/) (2019)

## Built on by (most cited first, 2 of 2)

- [Ascend HiFloat8 Format for Deep Learning](https://mlsanity.com/paper/2409.16626/) (2024)
- [PDPU: An Open-Source Posit Dot-Product Unit for Deep Learning Applications](https://mlsanity.com/paper/2302.01876/) (2023)

## Abstract (arXiv)

Low-precision formats have proven to be an efficient way to reduce not only the memory footprint but also the hardware resources and power consumption of deep learning computations. Under this premise, the posit numerical format appears to be a highly viable substitute for the IEEE floating-point, but its application to neural networks training still requires further research. Some preliminary results have shown that 8-bit (and even smaller) posits may be used for inference and 16-bit for training, while maintaining the model accuracy. The presented research aims to evaluate the feasibility to train deep convolutional neural networks using posits. For such purpose, a software framework was developed to use simulated posits and quires in end-to-end training and inference. This implementation allows using any bit size, configuration, and even mixed precision, suitable for different precision requirements in various stages. The obtained results suggest that 8-bit posits can substitute 32-bit floats during training with no negative impact on the resulting loss and accuracy.