Work

DiGamma: Domain-aware Genetic Algorithm for HW-Mapping Co-optimization for DNN Accelerators

Year: 2022

Type: conference-paper

Abstract: The design of DNN accelerators includes two key parts: HW resource configuration and mapping strategy. Intensive research has been conducted to optimize each of them independently. Unfortunately, optimizing for both together is extremely challenging due to the extremely large cross-coupled search space. To address this, in this paper, we propose a HW-Mapping co-optimization framework, an efficient encoding of the immense design space constructed by HW and Mapping, and a domain-aware genetic algorithm, named DiGamma, with specialized operators for improving search efficiency. We evaluate DiGamma with seven popular DNNs models with different properties. Our evaluations show DiGamma can achieve (geomean) 3.0x and 10.0x speedup, comparing to the best-performing baseline optimization algorithms, in edge and cloud settings.

Source: Design, Automation, and Test in Europe

DOI: 10.23919/date54114.2022.9774568

Authors: Sheng-Chun Kao, Michael Pellauer, Angshuman Parashar, Tushar Krishna

Institutions: Georgia Institute of Technology, Nvidia (United Kingdom), Nvidia (United States)

Language: English

Topic: Embedded Systems Design Techniques

Subfield: Hardware and Architecture

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: National Science Foundation

Award: OAC Core: SHF: Small: Enabling Rapid Design and Deployment of Deep Learning Models on Hardware Accelerators