CCIT seminar: on the Carbon Footprint of Deep Learning

Deep Learning (DL) has transformed several application domains, including computer vision and natural language processing, with new and exciting possibilities in Artificial Intelligence (AI). These advancements have been enabled, and accelerated, by large scale computations on massive data which also translate into increased energy- and carbon- costs. In this work, we take a look at the carbon footprint of DL across domains, present techniques to quantify it and practices that could improve the environmental sustainability of AI. Further, we will look at the notion of Pareto Sustainable AI which considers the three-way trade off between performance, environmental impact and factors impacting social sustainability (such as safety and fairness).