Impact of conductance drift on multi-PCM synaptic architectures

I. Boybat, S. R. Nandakumar, M. Le Gallo, B. Rajendran, Y. Leblebici, A. Sebastian, E. Eleftheriou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Scopus citations


In-memory computing with nanoscale memristive devices such as phase-change memory (PCM) has emerged as an alternative to conventional von Neumann systems to train deep neural networks (DNN) where a synaptic weight is represented by the device conductance. However, PCM devices exhibit temporal evolution of the conductance values referred to as the conductance drift, which poses challenges for maintaining synaptic weights reliably. Based on the mean behavior of 10,000 GST-based PCM devices, we observe that the drift coefficient is dependent on the conductance value. Moreover, we show that PCM drift is re-initialized and the drift history is erased after the application of even partial SET pulses. This is regardless of how much the device has drifted. With models capturing these features, we show that drift has a detrimental impact on training DNNs, but drift resilience can be significantly improved with a recently proposed multi-PCM synaptic architecture.

Original languageEnglish (US)
Title of host publicationNVMTS 2018 - Non-Volatile Memory Technology Symposium 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538677834
StatePublished - Jan 4 2019
Event18th Non-Volatile Memory Technology Symposium, NVMTS 2018 - Sendai, Japan
Duration: Oct 22 2018Oct 24 2018

Publication series

NameNVMTS 2018 - Non-Volatile Memory Technology Symposium 2018


Conference18th Non-Volatile Memory Technology Symposium, NVMTS 2018

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Electrical and Electronic Engineering
  • Safety, Risk, Reliability and Quality


  • deep neural network
  • drift
  • in-situ training
  • phase-change memory


Dive into the research topics of 'Impact of conductance drift on multi-PCM synaptic architectures'. Together they form a unique fingerprint.

Cite this