This project helped to build a globally scalable foundation model to enable near real-time tracking of climate change causes and impacts. A foundation model is a model (usually a deep neural network) that has been trained on a large and diverse set of data, after which it can be adapted to a variety of different (but related) inference tasks with a small fraction of additional training data and computation. Leveraging recent developments in self-supervised learning, the team developed the dataset and training paradigm for creating this foundation model and begin applying it to real-world data. A model developed using such a dataset will enhance climate change mitigation/adaptation monitoring and planning by developing robust features that can be used to monitor a broad range of climate change contributing factors (e.g. energy infrastructure and use, agricultural activity) and impacts (e.g. economic impacts and human migration) for informing climate mitigation and adaptation strategies.
Project Lead: Kyle Bradbury
Project Manager: Paul Markakis
View the team’s project poster here: Team 22
View the team’s video presentation here: