Short-term to long-term plasticity transition in filamentary switching for memory applications

Abstract

The emergence of memristive devices is currently driving an increasing interest in neuromorphic computing, which could complement and provide enhanced functionalities to existing CMOS/Von Neumann processors. Various plasticity mechanisms, analogous to synaptic plasticity in the brain, have indeed been implemented in emerging memristive systems. Additionally, we have recently demonstrated experimentally that several synaptic features can be embedded in a single memory component by exploiting the basic physics of filamentary resistive switching [1]. Here, by exploiting a memristive synaptic bio-model of this original behavior, we show how this device can modulate its weight in a Short-Term Plasticity (STP) to Long-Term Plasticity (LTP) transition, and how this can be harnessed in a neuromorphic memory application. These results pave the way for future engineering of neuromorphic computing systems, where complex behaviors of memristive physics can be exploited.