Scientist Wins Early Career Presidential Award

Duke computer scientist Vincent Conitzer has won a 2012 Presidential Early Career Award for Scientists and Engineers.

The award is the highest honor the federal government gives to young scientists and includes guaranteed funding for the winners for the next five years.

Conitzer was recognized specifically for his "groundbreaking work on computational aspects of game theory, social choice and mechanism design," according to a July 23 White House statement announcing the award winners.

"I am excited to win this award. If you look at the names of current and past winners, there are some amazing people receiving this recognition and being associated with them is an honor," said Conitzer, whose nomination for the award came from the Department of Defense's Army Research Office.

With the award, Conitzer said he plans to expand his research on game theory and how it can be applied to issues of security.

The work integrates the fundamentals of how games like rock-paper-scissors and poker are played to figure out the best strategy for problems such as placing air marshals on flights and protecting harbors and ports from attacks.

The mathematical principles that predict the outcomes of rock-paper-scissors and other single-round games, like Chicken, are "a good model to start with for developing systematic approaches" to security applications, Conitzer said. But he wants to look for the best solutions to a game as it evolves over time. These more advanced games, with fundamentals similar to poker, will provide a better understanding of how players might allocate their resources over time and how they learn their opponents' preferences and biases.

"In poker, a player places a bet. You have to infer what cards he might have and if he is bluffing and then make your own bet. Or, another card could be revealed, which also affects how you want to bet," Conitzer said. The situation evolves over time, which is more natural for modeling many security questions, and it is a feature scientists need to move toward in their models, he said.

Conitzer added that he also looks forward to using the award to continue to identify bright, motivated undergraduate and graduate students who want to work on this type of research. "I am so grateful to be in such a supportive environment in my department and at Duke. I have been given a lot of time to focus on research and integrate it with my teaching, which helps a lot," he said.

He explained that teaching courses related to his research exposes students to his ideas. Intrigued, many students then go on to work on independent projects with him after the courses end, which is exciting and enriching for him and for his students. "Duke students are extremely bright and talented and working with them greatly advances this line of research," Conitzer said. He specifically noted the contributions of Josh Letchford and Dima Korzhyk, two Ph.D. students currently working with him.

Conitzer joined Duke in 2006. He is the Sally Dalton Robinson Professor of Computer Science and a Professor of Economics. He completed his master's and doctoral degrees in computer science at Carnegie Mellon and earned his bachelor's degree in applied mathematics from Harvard. The National Science Foundation awarded Conitzer a Faculty Early Career Development, or CAREER, grant in 2010. He was listed among Artificial Intelligence's 10 to Watch by the Institute of Electrical and Electronics Engineers in 2011 and won the 2011 International Joint Conferences on Artificial Intelligence Computers and Thought Award.