Communication Dans Un Congrès Année : 2022

Voltage-to-Voltage Sigmoid Neuron Activation Function Design for Artificial Neural Networks

Résumé

An Artificial Neural Network (ANN) involves a complex network of interconnected nodes called artificial neurons (AN); the AN sums N weighted inputs and passes the result through a non-linear activation function (AF). In this work, a modified version of the sigmoid activation function is proposed. In order to obtain a voltage-to-voltage (V-V) transfer function required by our specific ANN, the proposed solution uses a pseudo-differential pair configuration at the input as voltage to current converter. The proposed circuit is designed in a 180nm CMOS technology of TSMC and is simulated in Cadence-Virtuoso for the proper transistor sizing in order to obtain the desired steepness of the sigmoid function. The simulations results show an overall a minimum error of 1.09 % compared to the mathematical function and power consumption of 6.77µW. Comparison with previous works and the actual mathematical function prove very favorable.

Fichier principal
Vignette du fichier
ML_LASCAS_2022_paper_65.pdf (583.4 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-03812298 , version 1 (12-10-2022)

Licence

Identifiants

Citer

Tatiana Moposita, Lionel Trojman, Felice Crupi, Marco Lanuzza, Andrei Vladimirescu. Voltage-to-Voltage Sigmoid Neuron Activation Function Design for Artificial Neural Networks. IEEE 13th Latin America Symposium on Circuits and System (LASCAS) 2022, Mar 2022, Puerto Varas, France. pp.1-4, ⟨10.1109/LASCAS53948.2022.9789075⟩. ⟨hal-03812298⟩

Collections

164 Consultations
949 Téléchargements

Altmetric

Partager

  • More