EUVIMEDEUVIMED entdecken
ÜbersichtsarbeitOpen Access · CC BY-NC-ND 4.0

Compute in‐Memory with Non‐Volatile Elements for Neural Networks: A Review from a Co‐Design Perspective

Wilfried Haensch, Anand Raghunathan, Kaushik Roy, Bhaswar Chakrabarti, Charudatta M Phatak, Cheng Wang, Supratik Guha iD

Advanced Materials · 2023-09 · 35 (37)

DOI: 10.1002/adma.202204944 ↗

Kurz gefasst

Übersichtsarbeit in der Zeitschrift Advanced Materials, online erschienen am 02.03.2023. Verfasst von Wilfried Haensch und 6 weiteren Autor:innen. Beteiligte Einrichtung: Materials Science Division, Argonne National Laboratory, Lemont, IL, 60439, USA. Die Arbeit wurde bisher 142-mal zitiert (Crossref) und nennt 220 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: U.S. Department of Energy, Office of Science.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep learning has become ubiquitous, touching daily lives across the globe. Today, traditional computer architectures are stressed to their limits in efficiently executing the growing complexity of data and models. Compute‐in‐memory (CIM) can potentially play an important role in developing efficient hardware solutions that reduce data movement from compute‐unit to memory, known as the von Neumann bottleneck. At its heart is a cross‐bar architecture with nodal non‐volatile‐memory elements that performs an analog multiply‐and‐accumulate operation, enabling the matrix‐vector‐multiplications repeatedly used in all neural network workloads. The memory materials can significantly influence final system‐level characteristics and chip performance, including speed, power, and classification accuracy. With an over‐arching co‐design viewpoint, this review assesses the use of cross‐bar based CIM for neural networks, connecting the material properties and the associated design constraints and demands to application, architecture, and performance. Both digital and analog memory are considered, assessing the status for training and inference, and providing metrics for the collective set of properties non‐volatile memory materials will need to demonstrate for a successful CIM technology.

Schlagwörter

Stichwörter der Autor:innen

analog computerscompute in-memorycross-bar arraysnon-volatile memory

Autor:innen und Institutionen

  • Materials Science Division, Argonne National Laboratory, Lemont, IL, 60439, USA.
  • Department of Electrical Engineering, Purdue University, West Lafayette, IN, 47907, USA.
  • Department of Electrical Engineering, Indian Institute of Technology Madras, Chennai, Tamil Nadu, 600036, India.
  • Pritzker School of Molecular Engineering, University of Chicago, Chicago, IL, 60637, USA.

ORCID-Profile: Supratik Guha

Förderung

  • U.S. Department of Energy
  • Office of Science

Zitieren

APA

Haensch, W., Raghunathan, A., Roy, K., Chakrabarti, B., Phatak, C. M., Wang, C., & Guha, S. (2023). Compute in‐Memory with Non‐Volatile Elements for Neural Networks: A Review from a Co‐Design Perspective. Advanced Materials, 35(37). https://doi.org/10.1002/adma.202204944

Vancouver

Haensch W, Raghunathan A, Roy K, Chakrabarti B, Phatak CM, Wang C, et al. Compute in‐Memory with Non‐Volatile Elements for Neural Networks: A Review from a Co‐Design Perspective. Advanced Materials. 2023;35(37). doi:10.1002/adma.202204944

BibTeX
@article{Haensch2023,
  author = {Haensch, Wilfried and Raghunathan, Anand and Roy, Kaushik and Chakrabarti, Bhaswar and Phatak, Charudatta M and Wang, Cheng and Guha, Supratik},
  title = {Compute in‐Memory with Non‐Volatile Elements for Neural Networks: A Review from a Co‐Design Perspective},
  journal = {Advanced Materials},
  year = {2023},
  volume = {35},
  number = {37},
  doi = {10.1002/adma.202204944}
}
Kostenlos starten →