“What we have the opportunity to do here is to transform engineering by taking this tool and using it in ways that haven’t yet been imagined,” said Andrea Goldsmith, dean of the School of Engineering and Applied Science and the Arthur LeGrand Doty Professor of Electrical and Computer Engineering at the launch event Aug. 29. “We have an opportunity to lead.”

Mengdi Wang, associate professor of electrical and computer engineering and the Center for Statistics and Machine Learning, and Ryan Adams, professor of computer science lead the initiative. Wang is also an associated faculty member of the Omenn-Darling Bioengineering Institute.

AI for Accelerating Invention is one of three research initiatives, including Natural and Artificial Minds and Princeton Language and Intelligence, which round out the Princeton Laboratory for Artificial Intelligence (AI Lab).

Using AI to push the frontier

The Aug. 29 event celebrated the launch of AI for Accelerated Invention with flash talks by ten Princeton engineering faculty members who showed how they use artificial intelligence as a tool in their research. Among the presenters were Omenn-Darling faculty members:

Cliff Brangwynne, June K. Wu '92 Professor of Chemical and Biological Engineering and director of the Omenn-Darling Bioengineering Institute, is researching the assembly of organelles – the specialized structures within living cells. Using neural network-based analyses of images and classification, his lab is developing approaches to understand the structures of organelles and how they function. Brangwynne’s lab is pioneering efforts to use this research to engineer synthetic organelles for biomedical and biotechnology applications. (Watch his presentation below.)

Kaushik Sengupta, associate professor of electrical and computer engineering, works on electronic and photonic circuits for wireless sensing and communication, including in smart biomedical devices. (Watch his presentation below.)

Jared Toettcher, associate professor of molecular biology and bioengineering and deputy director of the Omenn-Darling Bioengineering Institute, is working on artificial intelligence for probing and controlling biological systems. (Watch his presentation below.)

Ellen Zhong, assistant professor of computer science, and her colleagues developed software that uses machine learning to create 3D reconstructions of structures imaged by cryo-electron microscopy. These powerful microscopes produce millions of two-dimensional images that need to be carefully assembled to represent a three-dimensional image. Zhong’s software, CryoDRGN, uses a neural network to directly reconstruct the structures of molecules such as proteins, which carry out the functions necessary for biological life. With cryoDRGN mapping the 3D structures of these essential molecules, researchers could piece together a better understanding of proteins and the ways in which they function. (Watch her presentation below.)

“I was impressed by the breadth and depth of our engineering faculty, and the possibilities for using AI to push the frontier,” said Wang of the talks given at the AI^2 launch. “AI is not a stand-alone technology and we really need to work together to scale up things.”

“Researchers here at Princeton are tackling the most challenging and important scientific problems our society faces, with enthusiasm and creativity,” said Adams. “Our hope is that a collaborative community around artificial intelligence can amplify the capabilities of everyone at Princeton to achieve new scientific heights.”