Detail publikace

Evaluation of Memristor Models for Large Crossbar Structures

Originální název

Evaluation of Memristor Models for Large Crossbar Structures

Anglický název

Evaluation of Memristor Models for Large Crossbar Structures

Jazyk

en

Originální abstrakt

This paper is focused on comparing selected SPICE models of TiO2 memristors with respect to time- and memory requirements in the simulation of very large artificial neural networks, which are most likely the first real-world applications of memristors as analog memories. All models were implemented as HSPICE macros and simulated in a Multilayer Perceptron artificial neural network with variable configuration. The results show that after applying modifications to the models in order to prevent numerical overflows it is possible to simulate networks with tens of thousands of memristors.

Anglický abstrakt

This paper is focused on comparing selected SPICE models of TiO2 memristors with respect to time- and memory requirements in the simulation of very large artificial neural networks, which are most likely the first real-world applications of memristors as analog memories. All models were implemented as HSPICE macros and simulated in a Multilayer Perceptron artificial neural network with variable configuration. The results show that after applying modifications to the models in order to prevent numerical overflows it is possible to simulate networks with tens of thousands of memristors.

BibTex


@inproceedings{BUT128502,
  author="Zdeněk {Kolka} and Dalibor {Biolek} and Viera {Biolková} and Zdeněk {Biolek}",
  title="Evaluation of Memristor Models for Large Crossbar Structures",
  annote="This paper is focused on comparing selected SPICE models of TiO2 memristors with respect to time- and memory requirements in the simulation of very large artificial neural networks, which are most likely the first real-world applications of memristors as analog memories. All models were implemented as HSPICE macros and simulated in a Multilayer Perceptron artificial neural network with variable configuration. The results show that after applying modifications to the models in order to prevent numerical overflows it is possible to simulate networks with tens of thousands of memristors.",
  address="IEEE",
  booktitle="Proceedings of the 26th International Conference RADIOELEKTRONIKA 2016",
  chapter="128502",
  doi="10.1109/RADIOELEK.2016.7477423",
  howpublished="electronic, physical medium",
  institution="IEEE",
  year="2016",
  month="april",
  pages="91--94",
  publisher="IEEE",
  type="conference paper"
}