3 ms·
The Carbon Footprint of Machine Learning Training Will Plateau, Then Shrink
- belter 5y ago"...Machine Learning (ML) workloads have rapidly grown in importance, but raised concerns about their carbon footprint. Four best practices can reduce ML training energy by up to 100x and CO2 emissions up to 1000x. By following best practices, overall ML energy use (across research, development, and production) held steady at <15% of Google's total energy use for the past three years. If the whole ML field were to adopt best practices, total carbon emissions from training would reduce..." PDF: https://arxiv.org/ftp/arxiv/papers/2204/2204.05149.pdf https://arxiv.org/ftp/arxiv/papers/2204/2204.05149.pdf