Floating-floating point: a highly accurate number representation with flexible Counting ranges

TL;DR

This paper introduces Floating-Floating-Point (F2P), a flexible number representation that balances range and accuracy, improving performance in federated learning and network measurement applications.

Contribution

The paper proposes F2P, a novel floating-point system with adjustable mantissa and exponent partitioning, enhancing range and accuracy over traditional systems.

Findings

F2P improves network measurement accuracy.

F2P enhances federated learning performance.

F2P offers a larger counting range with better precision.

Abstract

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.