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

Abstract

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.

References in corpus (7)

• Deep Learning in Neural Networks: An Overview
• The Computational Limits of Deep Learning
• Rethinking floating point for deep learning
• Performance-Efficiency Trade-off of Low-Precision Numerical Formats in Deep Neural Networks
• Template-Based Posit Multiplication for Training and Inferring in Neural Networks
• Deep Learning Training on the Edge with Low-Precision Posits
• Training Deep Neural Networks Using Posit Number System

Cited by in corpus (2)

• PERCIVAL: Open-Source Posit RISC-V Core with Quire Capability
• PDPU: An Open-Source Posit Dot-Product Unit for Deep Learning Applications