Georgia Institute of Technology

Dec 11, 2017

Georgia Tech School of Computational Science and Engineering (CSE) Assistant Professor Bistra Dilkina has been awarded a grant from Microsoft as part of its AI for Earth program.

AI for Earth works to empower people and organizations to solve global environmental challenges by increasing access to artificial intelligence (AI) tools and educational opportunities, all while accelerating innovation.

"My research group focuses on developing machine learning and optimization techniques and applying them to challenging problems related to sustainable development goals, which closely aligns with Microsoft's AI for Earth program,” said Dilkina.

Via the Azure for Research AI for Earth award program, Microsoft provides selected researchers and organizations access to its cloud and AI computing resources to accelerate, improve, and expand work on climate change, agriculture, biodiversity, and water challenges.

The Georgia Tech project – known as Deep Population – is a collaborative effort with CSE Ph.D. students Caleb Robinson and Fred Hohman, as well as CSE Assistant Professor Polo Chau. Using Microsoft Azure infrastructure with Microsoft's CNTK deep-learning toolkit, the team will work to train and analyze deep convolutional neural networks that can predict the population of an area using only satellite imagery from the same area.

The team will use what they learn to develop a methodology that is applicable across the globe, including in areas that have no resources for expensive census surveys.

“We are particularly interested in human migration dynamics and would like to further use these models to study rural to urban migration trends and to document the effects of climate change,” said Dilkina.

“Microsoft’s generous grant will support our work on creating predictive models of human populations and migrations, enabling much faster progress and spatial extent for our analysis."

Georgia Tech is among the first grant recipients of AI for Earth, first launched in July 2017. The grant process was a competitive and selective process and was awarded in recognition of the potential of the work and power of AI to accelerate progress.

To date, Microsoft has distributed more than 35 grants to qualifying researchers and organizations around the world. It also recently announced its intent to put $50 million over five years into the program, enabling grant-making and educational training possible at a much larger scale.