A Practical Guide to Quantifying Carbon Emissions for Machine Learning researchers and practitioners

Anne-Laure Ligozat (1) , Sasha Luccioni

1 LISN - Laboratoire Interdisciplinaire des Sciences du Numérique

The goal of this short guide is to help the Machine Learning (ML) community better understand their carbon impact and to take steps to mitigate it. Carbon Tracking At the center of the climate crisis is a commonplace but very important concept: that of carbon dioxide (CO 2), low amounts of which occur naturally in the Earth's atmosphere, but its concentration has been rapidly increasing due to human activity. This increase is dangerous because of CO 2 's effect as a greenhouse gas, which contributes to global warming. It is therefore important to: 1) quantify the carbon impact of our actions; and 2) reduce, or mitigate, that impact in order to help slow down global warming and climate change more broadly.

Domaines: Informatique [cs]; Intelligence artificielle [cs.AI]; Apprentissage [cs.LG]

Anne-Laure Ligozat, Sasha Luccioni. A Practical Guide to Quantifying Carbon Emissions for Machine Learning researchers and practitioners. [Research Report] MILA; LISN. 2021. ⟨hal-03376391⟩