Delay Conditioned Generative Modelling of Resistive Drift in Memristors

TL;DR

This paper presents a novel generative modelling approach to simulate resistive drift in memristors, accounting for delay and initial resistance conditions, enhancing device reliability analysis for neuromorphic and storage applications.

Contribution

It introduces a data normalization scheme and a new training technique for conditioning generative models on continuous inputs, improving simulation efficiency and differentiability.

Findings

Effective simulation of resistive drift conditioned on device delay and initial resistance.

Enhanced model training through novel normalization and conditioning techniques.

Potential for improved device reliability assessment in neuromorphic computing.

Abstract

The modelling of memristive devices is an essential part of the development of novel in-memory computing systems. Models are needed to enable the accurate and efficient simulation of memristor device characteristics, for purposes of testing the performance of the devices or the feasibility of their use in future neuromorphic and in-memory computing architectures. The consideration of memristor non-idealities is an essential part of any modelling approach. The nature of the deviation of memristive devices from their initial state, particularly at ambient temperature and in the absence of a stimulating voltage, is of key interest, as it dictates their reliability as information storage media - a property that is of importance for both traditional storage and neuromorphic applications. In this paper, we investigate the use of a generative modelling approach for the simulation of the delay and initial resistance-conditioned resistive drift distribution of memristive devices. We introduce a data normalisation scheme and a novel training technique to enable the generative model to be conditioned on the continuous inputs. The proposed generative modelling approach is suited for use in end-to-end training and device modelling scenarios, including learned data storage applications, due to its simulation efficiency and differentiability.