Leveraging Bayesian Optimization to Speed Up Automatic Precision Tuning - Archive ouverte HAL Access content directly
Conference Papers Year :

Leveraging Bayesian Optimization to Speed Up Automatic Precision Tuning

(1) , (1)
1

Abstract

Using just the right amount of numerical precision is an important aspect for guaranteeing performance and energy efficiency requirements. Word-Length Optimization (WLO) is the automatic process for tuning the precision, i.e., bit-width, of variables and operations represented using fixed-point arithmetic. However, state-of-the-art precision tuning approaches do not scale well in large applications where many variables are involved. In this paper, we propose a hybrid algorithm combining Bayesian optimization (BO) and a fast local search to speed up the WLO procedure. Through experiments, we first show some evidence on how this combination can improve exploration time. Then, we propose an algorithm to automatically determine a reasonable transition point between the two algorithms. By statistically analyzing the convergence of the probabilistic models constructed during BO, we derive a stopping condition that determines when to switch to the local search phase. Experimental results indicate that our algorithm can reduce exploration time by up to 50%-80% for large benchmarks.
Fichier principal
Vignette du fichier
DATE2021_Hybrid_Approach.pdf (859.85 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03119548 , version 1 (24-01-2021)

Identifiers

  • HAL Id : hal-03119548 , version 1

Cite

Van-Phu Ha, Olivier Sentieys. Leveraging Bayesian Optimization to Speed Up Automatic Precision Tuning. DATE 2021 - 24th IEEE/ACM Design, Automation and Test in Europe, Feb 2021, Virtual Event, France. pp.1-6. ⟨hal-03119548⟩
106 View
238 Download

Share

Gmail Facebook Twitter LinkedIn More