mlsanityData to 7 Oct 2026

Accuracy is not the only Metric that matters: Estimating the Energy Consumption of Deep Learning Models

arXiv 2304.00897 ↗First posted 3 Apr 2023cs.LGcited by 9

Not on the map. Why?

Abstract (arXiv)

Modern machine learning models have started to consume incredible amounts of energy, thus incurring large carbon footprints (Strubell et al., 2019). To address this issue, we have created an energy estimation pipeline1, which allows practitioners to estimate the energy needs of their models in advance, without actually running or training them. We accomplished this, by collecting high-quality energy data and building a first baseline model, capable of predicting the energy consumption of DL models by accumulating their estimated layer-wise energies.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 11 of its 13 references that are in the map. First posted on arXiv 3 Apr 2023.

Builds on

11 of 13 references.

References · 11

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2019Energy and Policy Considerations for Deep Learning in NLPcited by 1,242
  4. 2019Quantifying the Carbon Emissions of Machine Learningcited by 516
  5. 2020Towards the Systematic Reporting of the Energy and Carbon Footprints of Machine Learningcited by 246
  6. 2020The Computational Limits of Deep Learningcited by 188
  7. 2021The Efficiency Misnomercited by 98
  8. 2017NeuralPower: Predict and Deploy Energy-Efficient Convolutional Neural Networkscited by 30
  9. 2016How to scale distributed deep learning?cited by 16
  10. 2021Dissecting FLOPs along input dimensions for GreenAI cost estimationscited by 10
  11. 2018Energy-based Tuning of Convolutional Neural Networks on Multi-GPUscited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 2; 2025: 5; 2026: 1.

9 papers.

  1. 2023Watt For What: Rethinking Deep Learning's Energy-Performance Relationshipcited by 16
  2. 2024Evaluating the Energy Consumption of Machine Learning: Systematic Literature Review and Experimentscited by 3
  3. 2025On the retraining frequency of global models in retail demand forecastingcited by 3
  4. 2024Normalizing Energy Consumption for Hardware-Independent Evaluationcited by 1
  5. 2025The cost of ensembling: is it always worth combining?cited by 1
  6. 2025An Efficient Dual-Line Decoder Network with Multi-Scale Convolutional Attention for Multi-organ Segmentationcited by 1
  7. 2025Dissecting Transformers: A CLEAR Perspective towards Green AIcited by 1
  8. 2025BIPPO: Budget-Aware Independent PPO for Energy-Efficient Federated Learning Servicescited by 0
  9. 2026WattLayer: Get Layers Right to Estimate Inference Energy of Neural Networkscited by 0