Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Evolution and Learning for Digital Circuit Design Alexander Nicholson
 

Summary: Evolution and Learning for Digital Circuit Design
Alexander Nicholson
Learning Systems Group
California Institute of Technology
136-93 Pasadena, CA, 91125
zander@work.caltech.edu
Abstract
We investigate the use of learning and evolu-
tion for digital hardware design. Using the
reactive tabu search for discrete optimiza-
tion, we show that we can learn a multiplier
circuit from a set of examples. The learned
circuit makes less than 2% error and uses
fewer chip resources than the standard digi-
tal design. We compare use of a genetic algo-
rithm and the reactive tabu search for tness
optimization. On a 2-bit adder design prob-
lem, the reactive tabu search performs signif-
icantly better for a similar execution time.
1 Introduction

  

Source: Abu-Mostafa, Yaser S. - Department of Mechanical Engineering & Computer Science Department, California Institute of Technology

 

Collections: Computer Technologies and Information Sciences