Log In Sign Up

The Carbon Footprint of Machine Learning Training Will Plateau, Then Shrink

by   David Patterson, et al.

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 the whole ML field were to adopt best practices, total carbon emissions from training would reduce. Hence, we recommend that ML papers include emissions explicitly to foster competition on more than just model quality. Estimates of emissions in papers that omitted them have been off 100x-100,000x, so publishing emissions has the added benefit of ensuring accurate accounting. Given the importance of climate change, we must get the numbers right to make certain that we work on its biggest challenges.


page 4

page 6


Preliminary Systematic Literature Review of Machine Learning System Development Process

Previous machine learning (ML) system development research suggests that...

Carbon Emissions and Large Neural Network Training

The computation demand for machine learning (ML) has grown rapidly recen...

The Values Encoded in Machine Learning Research

Machine learning (ML) currently exerts an outsized influence on the worl...

Carbontracker: Tracking and Predicting the Carbon Footprint of Training Deep Learning Models

Deep learning (DL) can achieve impressive results across a wide variety ...

Machine Learning Operations (MLOps): Overview, Definition, and Architecture

The final goal of all industrial machine learning (ML) projects is to de...