{"title":"Efficient Implementation of Tanh: A Comparative Study of New Results","authors":"Samira Sorayaasa, M. Ahmadi","doi":"10.5121/csit.2023.130701","DOIUrl":null,"url":null,"abstract":"Hyperbolic tangent (Tanh) activation function is used in multilayered artificial neural networks (ANN). This activation function contains exponential and division terms in its expressions which makes its accurate digital implementation difficult. In this paper we present two different approximation techniques for digital implementation of Tanh function using power of two and coordinate rotation digital computer (CORDIC) methods. A comparative study of both techniques in terms of accuracy of their approximations in hardware costs as well as their speed when implemented on FPGA is also explained","PeriodicalId":42597,"journal":{"name":"ADCAIJ-Advances in Distributed Computing and Artificial Intelligence Journal","volume":"68 1","pages":""},"PeriodicalIF":1.7000,"publicationDate":"2023-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"ADCAIJ-Advances in Distributed Computing and Artificial Intelligence Journal","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.5121/csit.2023.130701","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0
Abstract
Hyperbolic tangent (Tanh) activation function is used in multilayered artificial neural networks (ANN). This activation function contains exponential and division terms in its expressions which makes its accurate digital implementation difficult. In this paper we present two different approximation techniques for digital implementation of Tanh function using power of two and coordinate rotation digital computer (CORDIC) methods. A comparative study of both techniques in terms of accuracy of their approximations in hardware costs as well as their speed when implemented on FPGA is also explained