EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Using the IBM analog in-memory hardware acceleration kit for neural network training and inference

Manuel Le Gallo, Corey Lammie, Julian Büchel, Fabio Carta, Omobayode Fagbohungbe, Charles Mackin, Hsinyu Tsai, Vijay Narayanan, Abu Sebastian, Kaoutar El Maghraoui, Malte J. Rasch

APL Machine Learning · 01.12.2023 · 1 (4)

DOI: 10.1063/5.0168089 ↗

Kurz gefasst

Fachartikel in der Zeitschrift APL Machine Learning, online erschienen am 28.11.2023. Verfasst von Manuel Le Gallo und 10 weiteren Autor:innen. Die Arbeit wurde bisher 75-mal zitiert (Crossref) und nennt 80 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: HORIZON EUROPE European Innovation Council, Staatssekretariat für Bildung, Forschung und Innovation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Analog In-Memory Computing (AIMC) is a promising approach to reduce the latency and energy consumption of Deep Neural Network (DNN) inference and training. However, the noisy and non-linear device characteristics and the non-ideal peripheral circuitry in AIMC chips require adapting DNNs to be deployed on such hardware to achieve equivalent accuracy to digital computing. In this Tutorial, we provide a deep dive into how such adaptations can be achieved and evaluated using the recently released IBM Analog Hardware Acceleration Kit (AIHWKit), freely available at https://github.com/IBM/aihwkit. AIHWKit is a Python library that simulates inference and training of DNNs using AIMC. We present an in-depth description of the AIHWKit design, functionality, and best practices to properly perform inference and training. We also present an overview of the Analog AI Cloud Composer, a platform that provides the benefits of using the AIHWKit simulation in a fully managed cloud setting along with physical AIMC hardware access, freely available at https://aihw-composer.draco.res.ibm.com. Finally, we show examples of how users can expand and customize AIHWKit for their own needs. This Tutorial is accompanied by comprehensive Jupyter Notebook code examples that can be run using AIHWKit, which can be downloaded from https://github.com/IBM/aihwkit/tree/master/notebooks/tutorial.

Förderung

  • HORIZON EUROPE European Innovation Council
  • Staatssekretariat für Bildung, Forschung und Innovation

Zitieren

APA

Gallo, M. L., Lammie, C., Büchel, J., Carta, F., Fagbohungbe, O., Mackin, C., Tsai, H., Narayanan, V., Sebastian, A., Maghraoui, K. E., & Rasch, M. J. (2023). Using the IBM analog in-memory hardware acceleration kit for neural network training and inference. APL Machine Learning, 1(4). https://doi.org/10.1063/5.0168089

Vancouver

Gallo ML, Lammie C, Büchel J, Carta F, Fagbohungbe O, Mackin C, et al. Using the IBM analog in-memory hardware acceleration kit for neural network training and inference. APL Machine Learning. 2023;1(4). doi:10.1063/5.0168089

BibTeX
@article{Gallo2023,
  author = {Gallo, Manuel Le and Lammie, Corey and Büchel, Julian and Carta, Fabio and Fagbohungbe, Omobayode and Mackin, Charles and Tsai, Hsinyu and Narayanan, Vijay and Sebastian, Abu and Maghraoui, Kaoutar El and Rasch, Malte J.},
  title = {Using the IBM analog in-memory hardware acceleration kit for neural network training and inference},
  journal = {APL Machine Learning},
  year = {2023},
  volume = {1},
  number = {4},
  doi = {10.1063/5.0168089}
}
Kostenlos starten →