Summary of Computing Within Limits: An Empirical Study Of Energy Consumption in Ml Training and Inference, by Ioannis Mavromatis and Kostas Katsaros and Aftab Khan
Computing Within Limits: An Empirical Study of Energy Consumption in ML Training and Inference
by Ioannis Mavromatis, Kostas Katsaros, Aftab Khan
First submitted to arxiv on: 20 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper explores ways to reduce the environmental impact of machine learning (ML) by examining different models and hyperparameters in both training and inference phases. The authors use software-based power measurements to analyze various configurations, models, and datasets, identifying correlations between energy consumption and carbon footprint reduction while maintaining performance. The study provides practical guidelines for constructing sustainable ML operations, emphasizing energy consumption and carbon footprint reductions. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making machine learning more environmentally friendly. Right now, machine learning uses a lot of energy, which contributes to climate change. The researchers looked at different ways to train and use machine learning models to see how they affect the environment. They found some surprising things, like how short-lived profiling can help predict long-term energy consumption. This information can be used to make machine learning more sustainable and reduce its carbon footprint. |
Keywords
» Artificial intelligence » Inference » Machine learning
