Trends in AI inference energy consumption: Beyond the performance-vs-parameter laws of deep learning
Sustainable Computing: Informatics and Systems · 2023-04 · 38 · 100857
DOI: 10.1016/j.suscom.2023.100857 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sustainable Computing: Informatics and Systems, 2023-04. Verfasst von Radosvet Desislavov und 2 weiteren Autor:innen. Die Arbeit wurde bisher 196-mal zitiert (Crossref) und nennt 112 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Future of Life Institute, Horizon 2020 Framework Programme, Defense Advanced Research Projects Agency.
Abstract (Originalsprache)
The progress of some AI paradigms such as deep learning is said to be linked to an exponential growth in the number of parameters. There are many studies corroborating these trends, but does this translate into an exponential increase in energy consumption? In order to answer this question we focus on inference costs rather than training costs, as the former account for most of the computing effort, solely because of the multiplicative factors. Also, apart from algorithmic innovations, we account for more specific and powerful hardware (leading to higher FLOPS) that is usually accompanied with important energy efficiency optimisations. We also move the focus from the first implementation of a breakthrough paper towards the consolidated version of the techniques one or two year later. Under this distinctive and comprehensive perspective, we study relevant models in the areas of computer vision and natural language processing: for a sustained increase in performance we see a much softer growth in energy consumption than previously anticipated. The only caveat is, yet again, the multiplicative factor, as future AI increases penetration and becomes more pervasive.
Förderung
- Future of Life Institute
- Horizon 2020 Framework Programme
- Defense Advanced Research Projects Agency
- Horizon 2020
- España Ministerio de Ciencia e Innovación
- European Regional Development Fund
- Government of Valencia
Zitieren
APA
Desislavov, R., Martínez-Plumed, F., & Hernández-Orallo, J. (2023). Trends in AI inference energy consumption: Beyond the performance-vs-parameter laws of deep learning. Sustainable Computing: Informatics and Systems, 38, 100857. https://doi.org/10.1016/j.suscom.2023.100857
Vancouver
Desislavov R, Martínez-Plumed F, Hernández-Orallo J. Trends in AI inference energy consumption: Beyond the performance-vs-parameter laws of deep learning. Sustainable Computing: Informatics and Systems. 2023;38:100857. doi:10.1016/j.suscom.2023.100857
BibTeX
@article{Desislavov2023,
author = {Desislavov, Radosvet and Martínez-Plumed, Fernando and Hernández-Orallo, José},
title = {Trends in AI inference energy consumption: Beyond the performance-vs-parameter laws of deep learning},
journal = {Sustainable Computing: Informatics and Systems},
year = {2023},
volume = {38},
pages = {100857},
doi = {10.1016/j.suscom.2023.100857}
}
EUVIMED