English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Passive resistive random access memory (ReRAM) crossbar arrays, a promisingemerging technology used for analog matrix-vector multiplications, are farsuperior to their active (1T1R) counterparts in terms of the integrationdensity. However, current transfers of neural network weights into theconductance state of the memory devices in the crossbar architecture areaccompanied by significant losses in precision due to hardware variabilitiessuch as sneak path currents, biasing scheme effects and conductance tuningimprecision. In this work, training approaches that adapt techniques such asdropout, the reparametrization trick and regularization to TiO2 crossbarvariabilities are proposed in order to generate models that are better adaptedto their hardware transfers. The viability of this approach is demonstrated bycomparing the outputs and precision of the proposed hardware-aware network withthose of a regular fully connected network over a few thousand weight transfersusing the half moons dataset in a simulation based on experimental data. Forthe neural network trained using the proposed hardware-aware method, 79.5% ofthe test set's data points can be classified with an accuracy of 95% or higher,while only 18.5% of the test set's data points can be classified with thisaccuracy by the regularly trained neural network.

Hardware-aware Training Techniques for Improving Robustness of Ex-Situ Neural Network Transfer onto Passive TiO2 ReRAM Crossbars