Optimization of Projected Phase Change Memory for Analog In‐Memory Computing Inference

to related objects

Abstract: Phase change memory (PCM) is one of the most promising candidates for non‐von Neumann based analog in‐memory computing–particularly for inference of previously‐trained deep neural networks (DNN). It is shown that PCM electrical properties can be tuned systematically using a projection liner, which is designed for resistance drift mitigation, in the manufacturable mushroom PCM. A systematic study of the electrical properties‐including resistance values, memory window, resistance drift, read noise, and their impact on the accuracy of large neural networks of various types and with tens of millions of weights is performed. It is sown that the DNN accuracy can be improved by the PCM with liner for both the short term and long term after programming, due to reduced resistance drift and read noise, respectively, despite the trade‐off of reduced memory window. The liner conductance, PCM device characteristics, and network inference accuracy with PCM memory window and reset state conductance is correlated, which allows us to identify the device optimization space to achieve better short term and long term accuracy for large neural networks.

Location
Deutsche Nationalbibliothek Frankfurt am Main
Extent
Online-Ressource
Language
Englisch

Bibliographic citation
Optimization of Projected Phase Change Memory for Analog In‐Memory Computing Inference ; day:12 ; month:04 ; year:2023 ; extent:9
Advanced electronic materials ; (12.04.2023) (gesamt 9)

Creator
Li, Ning
Mackin, Charles
Chen, An
Brew, Kevin
Philip, Timothy
Simon, Andrew
Saraf, Iqbal
Han, Jin‐Ping
Ghazi Sarwat, Syed
Burr, Geoffrey W.
Rasch, Malte
Sebastian, Abu
Narayanan, Vijay
Saulnier, Nicole

DOI
10.1002/aelm.202201190
URN
urn:nbn:de:101:1-2023041315012152001615
Rights
Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
Last update
14.08.2025, 10:49 AM CEST

Data provider

This object is provided by:
Deutsche Nationalbibliothek. If you have any questions about the object, please contact the data provider.
Show original at data provider

Associated

  • Li, Ning
  • Mackin, Charles
  • Chen, An
  • Brew, Kevin
  • Philip, Timothy
  • Simon, Andrew
  • Saraf, Iqbal
  • Han, Jin‐Ping
  • Ghazi Sarwat, Syed
  • Burr, Geoffrey W.
  • Rasch, Malte
  • Sebastian, Abu
  • Narayanan, Vijay
  • Saulnier, Nicole

Other Objects (12)

Optimization of Projected Phase Change Memory for Analog In‐Memory Computing Inference (Adv. Electron. Mater. 6/2023)

Optimization of Projected Phase Change Memory for Analog In‐Memory Computing Inference (Adv. Electron. Mater. 6/2023)

Projected Mushroom Type Phase‐Change Memory

Projected Mushroom Type Phase‐Change Memory

In-Memory Interval Joins

In-Memory Interval Joins

In-memory mechanical computing

In-memory mechanical computing

In‐Memory Database Query

In‐Memory Database Query

In-Memory Interval Joins

In-Memory Interval Joins

Fast and robust analog in-memory deep neural network training

Fast and robust analog in-memory deep neural network training

Ultrathin Nitride Ferroic Memory with Large ON/OFF Ratios for Analog In‐memory Computing

Ultrathin Nitride Ferroic Memory with Large ON/OFF Ratios for Analog In‐memory Computing

Multifunctional In‐Memory Analog‐to‐Digital Converter for Next‐Gen Compute‐in‐Memory Systems

Multifunctional In‐Memory Analog‐to‐Digital Converter for Next‐Gen Compute‐in‐Memory Systems

Topological structure of the solution sets to neutral evolution inclusions driven by measures

Topological structure of the solution sets to neutral evolution inclusions driven by measures

Innovative Strategies for Photons Management on Ultrathin Silicon Solar Cells

Innovative Strategies for Photons Management on Ultrathin Silicon Solar Cells

Tree-based machine learning performed in-memory with memristive analog CAM

Tree-based machine learning performed in-memory with memristive analog CAM

Related objects