How to estimate carbon footprint when training deep learning models? A guide and review
Open accessThe DOI matches its Crossref record; 10 of the 10 references it cites were found in a registry or an index.
Abstract
Machine learning and deep learning models have become essential in the recent fast development of artificial intelligence in many sectors of the society. It is now widely acknowledge that the development of these models has an environmental cost that has been analyzed in many studies. Several online and software tools have been developed to track energy consumption while training machine learning models. In this paper, we propose a comprehensive introduction and comparison of these tools for AI practitioners wishing to start estimating the environmental impact of their work. We review the specific vocabulary, the technical requirements for each tool. We compare the energy consumption estimated by each tool on two deep neural networks for image processing and on different types of servers. From these experiments, we provide some advice for better choosing the right tool and infrastructure.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
Energy and Policy Considerations for Deep Learning in NLP
Strubell, E., Ganesh, A., & McCallum, A. (2019) · arXiv · doi:10.48550/arxiv.1906.02243 · Cited by 3,735
Energy and Policy Considerations for Deep Learning in NLP
Strubell, E., Ganesh, A., & McCallum, A. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1355 · Cited by 3,684 · cited 1×
“But it is also now widely documented that training and deploying deep learning projects has an impact on the environment [4, 5, 6, 7, 8, 9, 10, 11, 12, 13].”
BackgroundThe role of artificial intelligence in achieving the Sustainable Development Goals
Vinuesa, R., Azizpour, H., Leite, I., Balaam, M., Dignum, V., Domisch, S., Felländer, A., Langhans, S. D., Tegmark, M., & Fuso Nerini, F. (2020) · Nature Communications · doi:10.1038/s41467-019-14108-y · Cited by 3,294
Aligning artificial intelligence with climate change mitigation
Kaack, L. H., Donti, P. L., Strubell, E., Kamiya, G., Creutzig, F., & Rolnick, D. (2022) · Nature Climate Change · doi:10.1038/s41558-022-01377-7 · Cited by 570
Moro, A., & Lonza, L. (2018) · Transportation Research Part D: Transport and Environment · doi:10.1016/j.trd.2017.07.012 · Cited by 492 · cited 1×
“The default value is 447 gCO2eq/kWh (average carbon intensity value in gCO2eq/kWh in the EU in 2018 [36]).”
BackgroundSustainable AI: Environmental Implications, Challenges and Opportunities
Wu, C.-J., Raghavendra, R., Gupta, U., Acun, B., Ardalani, N., Maeng, K., Chang, G., Behram, F. A., Huang, J., Bai, C., et al. (2021) · arXiv · doi:10.48550/arxiv.2111.00364 · Cited by 405 · cited 2×
“As mentioned in previous studies [10, 14, 38], the energy consumption is underestimated, since all the tools only measure the consumption during training and not during deployment.”
BackgroundTackling climate change with machine learning
Rolnick, D., Donti, P. L., Kaack, L. H., Kochanski, K., Lacoste, A., Sankaran, K., Ross, A. S., Milojevic-Dupont, N., Jaques, N., Waldman-Brown, A., et al. (2023) · ACM Computing Surveys · doi:10.1145/3485128 · Cited by 336
Measuring the Carbon Intensity of AI in Cloud Instances
Dodge, J., Prewitt, T., Tachet des Combes, R., Odmark, E., Schwartz, R., Strubell, E., Luccioni, A. S., Smith, N. A., DeCario, N., & Buchanan, W. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533234 · Cited by 257 · cited 3×
“The recommendations complete previous works that intended to make machine learning researchers better understand their carbon impact and to take steps to mitigate it [17, 12].”
BackgroundResultEvaluating the carbon footprint of NLP methods: a survey and analysis of existing tools
Bannour, N., Ghannay, S., Névéol, A., & Ligozat, A.-L. (2021) · Proceedings of the Second Workshop on Simple and Efficient Natural Language Processing · doi:10.18653/v1/2021.sustainlp-1.2 · Cited by 98
Cited by
Green artificial intelligence initiatives: Potentials and challenges
Alzoubi, Y. I., & Mishra, A. (2024) · Journal of Cleaner Production · doi:10.1016/j.jclepro.2024.143090 · Cited by 206
Estimating the environmental impact of Generative-AI services using an LCA-based methodology
Berthelot, A., Caron, E., Jay, M., & Lefèvre, L. (2024) · Procedia CIRP · doi:10.1016/j.procir.2024.01.098 · Cited by 103
“As training is divided into steps constant in cost, measuring the electricity consumption of a few steps is enough to estimate the total cost of training [3], assuming knowledge of the original training.”
Salvagno, M., Cassai, A. D., Zorzi, S., Zaccarelli, M., Pasetto, M., Sterchele, E. D., Chumachenko, D., Gerli, A. G., Azamfirei, R., & Taccone, F. S. (2024) · PLOS ONE · doi:10.1371/journal.pone.0309208 · Cited by 32
“Lastly, attention must be paid to the energy consumption of AI systems and their consequent carbon footprint, which can be considerable, especially in the case of large-scale computational models [32].”
Background
Metadata
- Authors
- Bouza, L., Bugeau, A., & Lannelongue, L.
- Venue
- Environmental Research Communications
- Publisher
- IOP Publishing
- Published
- 2023-11-01
- Volume
- 5
- Issue
- 11
- Pages
- 115014
- ISSN
- 2515-7620 (ISSN-L)
- OpenAlex
- W4386532294
- Cited by
- 129
- References
- 10
- FWCI
- 9.46
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: How to estimate carbon footprint when training deep learning models? A guide and review. https://app.cochat.ai/papers/10.1088/2515-7620/acf81b
- The paper
- Bouza, L., Bugeau, A., & Lannelongue, L. (2023). How to estimate carbon footprint when training deep learning models? A guide and review. Environmental Research Communications. https://doi.org/10.1088/2515-7620/acf81b
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.