Using Phase-change memory as artificial synapses for neuromorphic computing hardware is severely limited by resistance drift. This drift depends on the initial resistance state, and devices show non-linear I-V characteristics indicating that the PCM resistance also depends on the read voltage. Here we present an electrical characterization methodology to asses these dependencies. We measure 180 reset cycles and perform statistical analysis. Our methodology can be used to better predict drift dependencies, ultimately mitigating it.
No takes yet. Share an insight, caveat, or question.
Nir-Harwood et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: