September 29, 2015

UW computer science alumnus wins a MacArthur Foundation ‘genius grant’

Computer scientist and University of Washington alumnus Christopher Ré is one of 24 recipients of “genius” grants this year from the John D. and Catherine T. MacArthur Foundation, the organization announced Sept. 28. Ré, an assistant professor of computer science at Stanford University, pursues new approaches to help computers analyze large, complex datasets. The diverse array of applications for Ré’s approach range from discovering new drugs to fighting human trafficking.

“He has taken the area of database research to a whole new level,” said Dan Suciu, UW professor of computer science and engineering and Ré’s former doctoral advisor. “I’m very happy for him.”

Under Suciu, Ré earned a Ph.D. in computer science and engineering at UW in 2009. The title of his dissertation — “Managing Large-scale Probabilistic Databases” — reveals his long-standing interest in how computers handle “big data,” particularly information scattered in formats that are difficult for computers to recognize, extract and scrutinize using traditional data analysis tools. One of Ré’s goals is to integrate and analyze this “unstructured” information to help solve practical problems.

As a MacArthur fellow, Ré will receive a stipend of $625,000, paid out in quarterly installments over the next five years. In a video interview with the MacArthur Foundation, Ré said the award is “one of the things you dream about,” especially since it provides resources to expand his research projects.

“We’re trying to summarize the vast information that’s out there — on all of the web pages, in all of the libraries, in all the government reports – in one place,” said Ré in the same video interview. “At least, that’s the dream.”

Ré has pursued innovative approaches to machine learning, statistical analysis and logic to help computers access and analyze “dark data,” the bits of data embedded in formats such as images, tables and texts. Dark data and other “unstructured” information are an untapped source of valuable information which, if properly accessed and analyzed by computers, could help fields from law enforcement to paleontology. Ré developed DeepDive, an inference engine which tries to sort through dark data, look for patterns, make inferences and compare them to existing datasets. For example, DeepDive has sorted through published scientific papers to extract data about the relationships among genes, drug therapies and diseases. It is also analyzing scientific records of fossil excavations to map all known sites of fossil finds. One program through DARPA is even using DeepDive to learn more about connections between groups engaged in human trafficking.

Ré is not the first MacArthur fellow associated with the UW Department of Computer Science & Engineering. Yoky Matsuoka, now vice president of technology and analytics at Twitter, was a professor in the department focused on neurorobotics research when she was named a MacArthur fellow in 2007. Current CSE and electrical engineering professor Shwetak Patel, who studies low-power sensing technology and leads the UW’s Ubiquitous Computing Lab, was named a fellow in 2011. UW biology professor and neuroscientist Tom Daniel, who is an adjunct CSE professor, was named a MacArthur fellow in 1996.