Follow Datanami:
July 13, 2021

Radiant Earth Foundation, Microsoft AI for Earth Partner for Sustainability Solutions

WASHINGTON, July 13, 2021 — The world’s ecological systems that humanity depends on are in disarray. To find solutions, a growing number of sustainability problem-solvers are now looking to artificial intelligence (AI) and Earth observation (EO) technologies. But these practitioners face many hurdles that hinder them from building and deploying practical AI applications at local and global levels. Today, Radiant Earth Foundation, a leading nonprofit dedicated to EO training data with machine learning (ML) algorithms, announces its partnership with Microsoft AI for Earth to support sustainability problem-solvers. The two organizations join forces to integrate their data ecosystems, Radiant MLHub and the Planetary Computer, which will give users access to better AI data, tools, and capacity development resources to accelerate solving sustainable development challenges using EO.

Microsoft AI for Earth, an initiative that puts AI tools and cloud computing in the hands of scientists, developers, and policymakers to solve sustainability challenges, launched the Planetary Computer to enable global-scale environmental monitoring. By partnering with them, Microsoft AI for Earth grantees will have the opportunity to publish their data and models on Radiant MLHub, the world’s first open-access repository for EO training data and models. With over 700 projects awarded in 80 countries, adding grantee data expands Radiant MLHub’s STAC-compliant datasets and model catalogs. It also adds to the map of training datasets that allow users and organizations to pinpoint geographical areas from which more data is needed. In addition, contributing to an open data ecosystem helps connect data sources and types, which is vital for contextualizing change.

Radiant Earth will also develop a set of workflows and tutorials to support users who would like to train scalable ML models and learn how to train or retrieve models on Radiant MLHub and the Planetary Computer. Furthermore, the two organizations will develop use cases and best practices towards establishing a training data on-demand service. This service will allow users to combine existing labeled data with new source data, which will enable efficient fusion of training datasets from different sources.

“We’re excited to bring our data ecosystem closer to the Planetary Computer,” said Hamed Alemohammad, Executive Director and Chief Data Scientist at Radiant Earth. “Microsoft’s notable investment in the sustainability space is galvanizing how we use technology to balance natural and socio-economic resources and mitigate impacts of climate change. This partnership further bridges the gap between the science and technology communities and allows them to efficiently deploy AI solutions for environmental sustainability at local and global scales.”

“The Planetary Computer already makes extensive use of the open standards and tools that have developed under the guidance of Radiant Earth, and we are thrilled to extend our collaboration to include Radiant MLHub as well,” added Dan Morris, Principal Scientist with the Microsoft AI for Earth program. “By combining Radiant MLHub’s library of training datasets with the Planetary Computer’s catalog of remote sensing and climate data, we aim to accelerate the development of ML tools for conservation and sustainability, and allow those tools to be applied at a global scale.”

About Radiant Earth

Radiant Earth Foundation is a nonprofit corporation working to empower organizations and individuals with open ML and Earth observation data, standards, and tools to address the world’s most critical international development challenges. Radiant Earth fosters collaboration through a cloud-based open geospatial training data library, Radiant MLHub. Radiant also supports an ecosystem of practitioners to develop standards, expand interoperability around ML on Earth observation, and provide information and training to help advance the capacity of those working in the global sector using ML and Earth observation.

About Microsoft AI for Earth

The Microsoft AI for Earth program supports organizations working at the intersection of artificial intelligence and environmental sustainability, by providing grants, open-source machine learning tools, and the Planetary Computer, which puts large-scale geospatial data in the hands of sustainability practitioners.


Source: Radiant Earth

Datanami