Publication

Accuracy Booster: Enabling 4-bit Fixed-point Arithmetic for DNN Training

2022

Abstract

The unprecedented demand for computing resources to train DNN models has led to a search for minimal numerical encoding. Recent state-of-the-art (SOTA) proposals advocate for multi-level scaled narrow bitwidth numerical formats. In this paper, we show that single-level scaling is sufficient to maintain training accuracy while maximizing arithmetic density. We identify a previously proposed single-level scaled format for 8-bit training, Hybrid Block Floating Point (HBFP), as the optimal candidate to minimize. We perform a full-scale exploration of the HBFP design space using mathematical tools to study the interplay among various parameters and identify opportunities for even smaller encodings across layers and epochs. Based on our findings, we propose Accuracy Booster, a mixedmantissa HBFP technique that uses 4-bit mantissas for over 99% of all arithmetic operations in training and 6-bit mantissas only in the last epoch and first/last layers. We show Accuracy Booster enables increasing arithmetic density over all other SOTA formats by at least 2.3× while achieving state-of-the-art accuracies in 4-bit training.

Official source

https://infoscience.epfl.ch/entities/publication/e7532245-2fa8-4bee-92cf-6bec0ffb688c

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Graph Chat

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

2022

Abstract

Official source

https://infoscience.epfl.ch/entities/publication/e7532245-2fa8-4bee-92cf-6bec0ffb688c

About this result

Related concepts (32)

8-bit computing

In computer architecture, 8-bit integers or other data units are those that are 8 bits wide (1 octet). Also, 8-bit central processing unit (CPU) and arithmetic logic unit (ALU) architectures are those that are based on registers or data buses of that size. Memory addresses (and thus address buses) for 8-bit CPUs are generally larger than 8-bit, usually 16-bit. 8-bit microcomputers are microcomputers that use 8-bit microprocessors.

Floating-point arithmetic

In computing, floating-point arithmetic (FP) is arithmetic that represents subsets of real numbers using an integer with a fixed precision, called the significand, scaled by an integer exponent of a fixed base. Numbers of this form are called floating-point numbers. For example, 12.345 is a floating-point number in base ten with five digits of precision: However, unlike 12.345, 12.3456 is not a floating-point number in base ten with five digits of precision—it needs six digits of precision; the nearest floating-point number with only five digits is 12.

Minifloat

In computing, minifloats are floating-point values represented with very few bits. Predictably, they are not well suited for general-purpose numerical calculations. They are used for special purposes, most often in computer graphics, where iterations are small and precision has aesthetic effects. Machine learning also uses similar formats like bfloat16. Additionally, they are frequently encountered as a pedagogical tool in computer-science courses to demonstrate the properties and structures of floating-point arithmetic and IEEE 754 numbers.