Publications

IrEne: Interpretable Energy Prediction for Transformers

Transformer Models Energy prediction

Authors: Qingqing Cao, Yash Kumar Lal, Harsh Trivedi, Aruna Balasubramanian, Niranjan Balasubramanian

Year: 2021

Published in: IJCNLP.

Read me: DOI: 10.18653/v1/2021.acl-long.167. 👩‍💻Replication package.

Abstract: Existing software-based energy measurements of NLP models are not accurate because they do not consider the complex interactions between energy consumption and model execution. We present IrEne, an interpretable and extensible energy prediction system that accurately predicts the inference energy consumption of a wide range of Transformer-based NLP models. IrEne constructs a model tree graph that breaks down the NLP model into modules that are further broken down into low-level machine learning (ML) primitives. IrEne predicts the inference energy consumption of the ML primitives as a function of generalizable features and fine-grained runtime resource usage. IrEne then aggregates these low-level predictions recursively to predict the energy of each module and finally of the entire model. Experiments across multiple Transformer models show IrEne predicts inference energy consumption of transformer models with an error of under 7% compared to the ground truth. In contrast, existing energy models see an error of over 50%. We also show how IrEne can be used to conduct energy bottleneck analysis and to easily evaluate the energy impact of different architectural choices. We release the code and data at https://github.com/StonyBrookNLP/irene.

@inproceedings{DBLP:conf/acl/CaoLTBB20,
  author       = {Qingqing Cao and
                Yash Kumar Lal and
                Harsh Trivedi and
                Aruna Balasubramanian and
                Niranjan Balasubramanian},
  editor       = {Chengqing Zong and
                Fei Xia and
                Wenjie Li and
                Roberto Navigli},
  title        = {IrEne: Interpretable Energy Prediction for Transformers},
  booktitle    = {Proceedings of the 59th Annual Meeting of the Association for Computational
                Linguistics and the 11th International Joint Conference on Natural
                Language Processing, {ACL/IJCNLP} 2021, (Volume 1: Long Papers), Virtual
                Event, August 1-6, 2021},
  pages        = {2145--2157},
  publisher    = {Association for Computational Linguistics},
  year         = {2021},
  url          = {https://doi.org/10.18653/v1/2021.acl-long.167},
  doi          = {10.18653/V1/2021.ACL-LONG.167},
  timestamp    = {Mon, 09 Aug 2021 16:25:37 +0200},
  biburl       = {https://dblp.org/rec/conf/acl/CaoLTBB20.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
  }