Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cohen, Itamar, Einziger, Gil
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2410.03692
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914965248540672
author Cohen, Itamar
Einziger, Gil
author_facet Cohen, Itamar
Einziger, Gil
contents Efficient number representation is essential for federated learning, natural language processing, and network measurement solutions. Due to timing, area, and power constraints, such applications use narrow bit-width (e.g., 8-bit) number systems. The widely used floating-point systems exhibit a trade-off between the counting range and accuracy. This paper introduces Floating-Floating-Point (F2P) - a floating point number that varies the partition between mantissa and exponent. Such flexibility leads to a large counting range combined with improved accuracy over a selected sub-range. Our evaluation demonstrates that moving to F2P from the state-of-the-art improves network measurement accuracy and federated learning.
format Preprint
id arxiv_https___arxiv_org_abs_2410_03692
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Floating-floating point: a highly accurate number representation with flexible Counting ranges
Cohen, Itamar
Einziger, Gil
Networking and Internet Architecture
Machine Learning
Efficient number representation is essential for federated learning, natural language processing, and network measurement solutions. Due to timing, area, and power constraints, such applications use narrow bit-width (e.g., 8-bit) number systems. The widely used floating-point systems exhibit a trade-off between the counting range and accuracy. This paper introduces Floating-Floating-Point (F2P) - a floating point number that varies the partition between mantissa and exponent. Such flexibility leads to a large counting range combined with improved accuracy over a selected sub-range. Our evaluation demonstrates that moving to F2P from the state-of-the-art improves network measurement accuracy and federated learning.
title Floating-floating point: a highly accurate number representation with flexible Counting ranges
topic Networking and Internet Architecture
Machine Learning
url https://arxiv.org/abs/2410.03692