Toward Software-Equivalent Accuracy on Transformer-Based Deep Neural Networks With Analog Memory Devices.

BERT DNN PCM RRAM Transformer analog accelerators in-memory computing

Journal

Frontiers in computational neuroscience
ISSN: 1662-5188
Titre abrégé: Front Comput Neurosci
Pays: Switzerland
ID NLM: 101477956

Informations de publication

Date de publication:
2021
Historique:
received: 03 03 2021
accepted: 14 05 2021
entrez: 22 7 2021
pubmed: 23 7 2021
medline: 23 7 2021
Statut: epublish

Résumé

Recent advances in deep learning have been driven by ever-increasing model sizes, with networks growing to millions or even billions of parameters. Such enormous models call for fast and energy-efficient hardware accelerators. We study the potential of Analog AI accelerators based on Non-Volatile Memory, in particular Phase Change Memory (PCM), for software-equivalent accurate inference of natural language processing applications. We demonstrate a path to software-equivalent accuracy for the GLUE benchmark on BERT (Bidirectional Encoder Representations from Transformers), by combining noise-aware training to combat inherent PCM drift and noise sources, together with reduced-precision digital attention-block computation down to INT6.

Identifiants

pubmed: 34290595
doi: 10.3389/fncom.2021.675741
pmc: PMC8287521
doi:

Types de publication

Journal Article

Langues

eng

Pagination

675741

Informations de copyright

Copyright © 2021 Spoon, Tsai, Chen, Rasch, Ambrogio, Mackin, Fasoli, Friz, Narayanan, Stanisavljevic and Burr.

Déclaration de conflit d'intérêts

The authors were employed by IBM Research.

Références

Nat Nanotechnol. 2015 Mar;10(3):209-20
pubmed: 25740132
Nature. 2015 May 28;521(7553):436-44
pubmed: 26017442
Nature. 2018 Jun;558(7708):60-67
pubmed: 29875487
Nat Commun. 2020 May 18;11(1):2473
pubmed: 32424184

Auteurs

Katie Spoon (K)

IBM Research-Almaden, San Jose, CA, United States.

Hsinyu Tsai (H)

IBM Research-Almaden, San Jose, CA, United States.

An Chen (A)

IBM Research-Almaden, San Jose, CA, United States.

Malte J Rasch (MJ)

IBM T. J. Watson Research Center, Yorktown Heights, NY, United States.

Stefano Ambrogio (S)

IBM Research-Almaden, San Jose, CA, United States.

Charles Mackin (C)

IBM Research-Almaden, San Jose, CA, United States.

Andrea Fasoli (A)

IBM Research-Almaden, San Jose, CA, United States.

Alexander M Friz (AM)

IBM Research-Almaden, San Jose, CA, United States.

Pritish Narayanan (P)

IBM Research-Almaden, San Jose, CA, United States.

Milos Stanisavljevic (M)

IBM Zurich Research Center, Zurich, Switzerland.

Geoffrey W Burr (GW)

IBM Research-Almaden, San Jose, CA, United States.

Classifications MeSH