An efficient initialization mechanism of neurons for winner takes all neural network implemented in the CMOS technology

DOI10.1016/J.AMC.2015.04.123MaRDI QIDQ669659zbMATH OpenOpenAlexFDO

Authors Tomasz Talaśka, Marta Kolasa, Rafał Długosz, Pierre-André Farine

Publication date 15 March 2019

Published in Applied Mathematics and Computation (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.amc.2015.04.123

zbMATH Keywords

analog circuitsCMOS implementationconvex combination methodinitialization mechanismlow energy consumptionwinner takes all neural network

Mathematics Subject Classification ID

Neural networks for/in biological studies, artificial life and related topics (92B20)

Recommendations

• Efficient methods of initializing neuron weights in self-organizing networks implemented in hardware
• A novel recursive algorithm used to model hardware programmable neighborhood mechanism of self-organizing neural networks
• scientific article; zbMATH DE number 1928651
• scientific article; zbMATH DE number 2050894
• Study of the Kohonen network with a discrete state space

Cites work

• scientific article; zbMATH DE number 1455125 (Why is no real title available?)
• Principles of Artificial Neural Networks
• Self-organizing maps.

Cited in

(3)

• On a constant-time, low-complexity winner-take-all neural network
• A novel, flexible circuit used to implement selected mathematical operations for AI algorithms optimized for hardware applications
• Efficient methods of initializing neuron weights in self-organizing networks implemented in hardware

This page was built for publication: An efficient initialization mechanism of neurons for winner takes all neural network implemented in the CMOS technology

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q669659)