Climate models are essential for policy and decision makers. But running a climate model at city-resolution (1km) takes multiple weeks on a 5000x GPU supercomputer; consuming the same electricity a coal power plant would generate in one hour. As a result city-scale information of climate risks is inaccessible. This talk presents a pathway how machine learning can create copies of climate models that run in seconds rather than weeks.
Activism, AI, Climate Change, Machine Learning, Policy, Social Justice Björn Lütjens is a PhD Candidate at MIT. His research is tackling climate change with machine learning, little-by-little, together with Prof. D. Newman, C. Crawford, and C. Hill. He aims to break down access barriers to city-scale forecasts of climate risks--using machine learning and physics-based models. He also crochets, windsurfs poorly, and loves meeting new people--you're no exception--please don't hesitate to reach out at lutjens at mit [dot] edu. This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at [ Ссылка ]
