Machine Learning Can Help Create a Sustainable Future

From rapidly identifying mosquito species to solving the world’s most pressing problems, machine learning algorithms will play an important role in building a sustainable future.

At least, that’s how Microsoft’s Chief Environmental Officer Lucas Joppa sees it.

Joppa spent an hour last week (Nov. 4) speaking to an online gathering of Yale School of Public Health students and faculty members about machine learning’s potential — especially when it comes to tracking diseases. In fact, he said, self-training computer programs can even help fill major gaps in our understanding of environmental systems large and small.

"When it comes to the natural world, we know a lot about a little and a little about a lot. That’s pretty clear,” he said. “There’s a big opportunity, I think, for the technology sector.”

And it’s already happening. Joppa heads Microsoft’s AI for Earth program, a five year, multi-million-dollar initiative dedicated to using artificial intelligence for the global good.

His program also has significant impacts for public health, he said.

Joppa took his audience through an early project that the initiative funded. Scientists wanted to learn more about the viruses circling throughout an ecosystem by capturing and analyzing the blood found in the abdomens of hungry mosquitoes. They set up smart traps that were able to identify the species of mosquito within milliseconds thanks to machine learning. Later, the team could pick out the viruses’ genetic code with surprising accuracy.

These kinds of projects, Joppa explained, “really change the way that we peer into these natural worlds.”

“We’re going to have to understand [them] much better if we are going to be able to put — and then keep healthy — 10 billion people on planet Earth,” he added.

Other applications abound. Artificial intelligence can help avert water crises and can improve the accuracy of air quality measurements. All told, Joppa’s initiative gives out thousands of dollars in grants and computer time to partners across the globe — and will continue to do so for at least the next several years.

“I kind of see it as my job — and as Microsoft’s job — to produce and provide the building blocks that make all of those organizations work as effective and scalable as possible,” he said. “The world needs a planetary computer.”

To Brady Rowe, a YSPH student who attended the virtual talk, Joppa’s initiative could be an opportunity to explore ideas related to rotational cattle farming.

“I can write it down on paper all day long,” Rowe asked, “but if I wanted to bring it out into the world, what’s a good first step for this kind of thing?”

Joppa, Ph.D., encouraged Rowe to apply for a grant. And if that fails, Joppa said he’d help introduce Rowe to others in the field.

“That’s one of the things we want to be as good as possible at is making those connections between like-minded researchers,” he added.

Joppa’s lecture is the latest in the Yale School of Public Health’s Department of Environmental Health Sciences seminar series.