Evolutionary Algorithms in Approximate Computing: A Survey

TL;DR

This survey reviews how evolutionary algorithms are used in approximate computing, highlighting their roles in multi-objective optimization, system parameter tuning, and architecture design, including emerging neural network applications.

Contribution

It provides the first comprehensive overview of EA-based approaches in approximate computing, categorizing methods and identifying new research directions.

Findings

EAs are mainly used as multi-objective optimizers.

Applications span all design abstraction levels.

Neural architecture search is an emerging area.

Abstract

In recent years, many design automation methods have been developed to routinely create approximate implementations of circuits and programs that show excellent trade-offs between the quality of output and required resources. This paper deals with evolutionary approximation as one of the popular approximation methods. The paper provides the first survey of evolutionary algorithm (EA)-based approaches applied in the context of approximate computing. The survey reveals that EAs are primarily applied as multi-objective optimizers. We propose to divide these approaches into two main classes: (i) parameter optimization in which the EA optimizes a vector of system parameters, and (ii) synthesis and optimization in which EA is responsible for determining the architecture and parameters of the resulting system. The evolutionary approximation has been applied at all levels of design abstraction and in many different applications. The neural architecture search enabling the automated hardware-aware design of approximate deep neural networks was identified as a newly emerging topic in this area.