Fixed-Posit: A Floating-Point Representation for Error-Resilient Applications

04/10/2021
by   Varun Gohil, et al.
0

Today, almost all computer systems use IEEE-754 floating point to represent real numbers. Recently, posit was proposed as an alternative to IEEE-754 floating point as it has better accuracy and a larger dynamic range. The configurable nature of posit, with varying number of regime and exponent bits, has acted as a deterrent to its adoption. To overcome this shortcoming, we propose fixed-posit representation where the number of regime and exponent bits are fixed, and present the design of a fixed-posit multiplier. We evaluate the fixed-posit multiplier on error-resilient applications of AxBench and OpenBLAS benchmarks as well as neural networks. The proposed fixed-posit multiplier has 47 posit multipliers and up to 70 respectively when compared to 32-bit IEEE-754 multiplier. These savings are accompanied with minimal output quality loss (1.2 across OpenBLAS and AxBench workloads. Further, for neural networks like ResNet-18 on ImageNet we observe a negligible accuracy loss (0.12 the fixed-posit multiplier.

READ FULL TEXT
research
05/18/2023

Comparative Study: Standalone IEEE 16-bit Floating-Point for Image Classification

Reducing the number of bits needed to encode the weights and activations...
research
03/29/2021

Representation range needs for 16-bit neural network training

Deep learning has grown rapidly thanks to its state-of-the-art performan...
research
05/30/2021

A Tapered Floating Point Extension for the Redundant Signed Radix 2 System Using the Canonical Recoding

A tapered floating point encoding is proposed which uses the redundant s...
research
01/05/2021

An Investigation on Inherent Robustness of Posit Data Representation

As the dimensions and operating voltages of computer electronics shrink ...
research
08/30/2020

Floating-Point Multiplication Using Neuromorphic Computing

Neuromorphic computing describes the use of VLSI systems to mimic neuro-...
research
07/12/2019

Posit NPB: Assessing the Precision Improvement in HPC Scientific Applications

Floating-point operations can significantly impact the accuracy and perf...
research
08/05/2019

PERI: A Posit Enabled RISC-V Core

Owing to the failure of Dennard's scaling the last decade has seen a ste...

Please sign up or login with your details

Forgot password? Click here to reset