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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Emerging non-volatile memories have been proposed for a wide range ofapplications from easing the von-Neumann bottleneck to neuromorphicapplications. Specifically, scalable RRAMs based on Pr$_{1-x}$Ca$_x$MnO$_3$(PCMO) exhibit analog switching have been demonstrated as an integratingneuron, an analog synapse, and a voltage-controlled oscillator. More recently,the inherent stochasticity of memristors has been proposed for efficienthardware implementations of Boltzmann Machines. However, as the problem sizescales, the number of neurons increase and controlling the stochasticdistribution tightly over many iterations is necessary. This requiresparametric control over stochasticity. Here, we characterize the stochastic Setin PCMO RRAMs. We identify that the Set time distribution depends on theinternal state of the device (i.e., resistance) in addition to external input(i.e., voltage pulse). This requires the confluence of contradictory propertieslike stochastic switching as well as deterministic state control in the samedevice. Unlike, "stochastic-everywhere" filamentary memristors, in PCMO RRAMs,we leverage the (i) stochastic Set in negative polarity and (ii) deterministicanalog Reset in positive polarity to demonstrate 100x reduced Set timedistribution drift. The impact on Boltzmann Machines' performance is analyzedand as opposed to the "fixed external input stochasticity", the"state-monitored stochasticity" can solve problems 20x larger in size. Statemonitoring also tunes out the device-to-device variability effect ondistributions providing 10x better performance. In addition to the physicalinsights, this study establishes the use of experimental stochasticity in PCMORRAMs in stochastic recurrent neural networks reliably over many iterations.

Stochasticity Invariance Control in Pr$_{1-x}$Ca$_x$MnO$_3$ RRAM to enable Large-Scale Stochastic Recurrent Neural Networks