CUED Publications database

A case for efficient accelerator design space exploration via Bayesian optimization

Reagen, B and Hernandez-Lobato, JM and Adolf, R and Gelbart, M and Whatmough, P and Wei, GY and Brooks, D (2017) A case for efficient accelerator design space exploration via Bayesian optimization. In: ACM/IEEE International Symposium on Low Power Electronics and Design, 2017-7-24 to 2017-7-26, Taipei, Taiwan.

Full text not available from this repository.


In this paper we propose using machine learning to improve the design of deep neural network hardware accelerators. We show how to adapt multi-objective Bayesian optimization to overcome a challenging design problem: Optimizing deep neural network hardware accelerators for both accuracy and energy efficiency. DNN accelerators exhibit all aspects of a challenging optimization space: The landscape is rough, evaluating designs is expensive, the objectives compete with each other, and both design spaces (algorithmic and microarchitectural) are unwieldy. With multi-objective Bayesian optimization, the design space exploration is made tractable and the design points found vastly outperform traditional methods across all metrics of interest.

Item Type: Conference or Workshop Item (UNSPECIFIED)
Uncontrolled Keywords: accelerator design space exploration bayesian optimization machine learning deep neural network hardware accelerators DNN accelerators optimization space
Divisions: Div F > Computational and Biological Learning
Depositing User: Cron Job
Date Deposited: 24 Oct 2017 01:42
Last Modified: 10 Apr 2021 00:49
DOI: 10.1109/ISLPED.2017.8009208