# Performance-Efficiency Trade-off of Low-Precision Numerical Formats in Deep Neural Networks

arXiv: https://arxiv.org/abs/1903.10584 · HTML: https://mlsanity.com/paper/1903.10584/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Mar 2019.

## Built on by (most cited first, 7 of 7)

- [EvGNN: An Event-driven Graph Neural Network Accelerator for Edge Vision](https://mlsanity.com/paper/2404.19489/) (2024)
- [Resource-Efficient Deep Learning: A Survey on Model-, Arithmetic-, and Implementation-Level Techniques](https://mlsanity.com/paper/2112.15131/) (2021)
- [Number Systems for Deep Neural Network Architectures: A Survey](https://mlsanity.com/paper/2307.05035/) (2023)
- [Low-Precision Mixed-Computation Models for Inference on Edge](https://mlsanity.com/paper/2312.02210/) (2023)
- [PositNN: Training Deep Neural Networks with Mixed Low-Precision Posit](https://mlsanity.com/paper/2105.00053/) (2021)
- [Energy-efficient Task Adaptation for NLP Edge Inference Leveraging Heterogeneous Memory Architectures](https://mlsanity.com/paper/2303.16100/) (2023)
- [Error Diffusion: Post Training Quantization with Block-Scaled Number Formats for Neural Networks](https://mlsanity.com/paper/2410.11203/) (2024)

## Abstract (arXiv)

Deep neural networks (DNNs) have been demonstrated as effective prognostic models across various domains, e.g. natural language processing, computer vision, and genomics. However, modern-day DNNs demand high compute and memory storage for executing any reasonably complex task. To optimize the inference time and alleviate the power consumption of these networks, DNN accelerators with low-precision representations of data and DNN parameters are being actively studied. An interesting research question is in how low-precision networks can be ported to edge-devices with similar performance as high-precision networks. In this work, we employ the fixed-point, floating point, and posit numerical formats at ≤8-bit precision within a DNN accelerator, Deep Positron, with exact multiply-and-accumulate (EMAC) units for inference. A unified analysis quantifies the trade-offs between overall network efficiency and performance across five classification tasks. Our results indicate that posits are a natural fit for DNN inference, outperforming at ≤8-bit precision, and can be realized with competitive resource requirements relative to those of floating point.