Exponential decreases in computation and communication costs induce marvelous new Internet services and social opportunities. However, new forms of social engagement also create new problems. This Integrative Graduate Education and Research Traineeship (IGERT) renewal will advance a principled, multi-disciplinary research method well-suited to make the Internet a place that is safe, fun and sustainable, with corresponding potential for improved social, economic, and political interaction. Incentive-centered design is a science that aligns participants' incentives with system or social goals. Distributed and collaborative system performance depends critically on strategic choices that users make when interacting with the system or each other, yet mismatch between individual interests and system goals is pervasive. This program takes a broad view of individual motivations, drawing on economic, psychological, and sociological theories, and combines these with the design and engineering sciences of artificial intelligence, software and networking.
The broader impacts go beyond research contributions to the design of socially-valuable Internet communities and services. The joint Ph.D. program at the University of Michigan and Wayne State University (a metropolitan comprehensive institution) will train future scholars and teachers from a diverse array of socio-economic and cultural backgrounds. In addition it includes a summer program for undergraduates from underrepresented groups, who in teams together with the IGERT trainees will develop submissions to international research competitions. This program will increase the pool of students from underrepresented groups who are prepared for and motivated to pursue graduate education. IGERT is an NSF-wide program intended to meet the challenges of educating U.S. Ph.D. scientists and engineers with the interdisciplinary background, deep knowledge in a chosen discipline, and the technical, professional, and personal skills needed for the career demands of the future. The program is intended to catalyze a cultural change in graduate education by establishing innovative new models for graduate education and training in a fertile environment for collaborative research that transcends traditional disciplinary boundaries.
The IGERT support in the past six years has provided our faculty and students with many opportunities for formal (course work, interdisciplinary seminar series) and informal interactions (research workshops, orientation parties). Consequently, we have produced a body of research which contributes to the development of new research areas at the intersection of (1) computer science and economics, (2) computer science and political science, as well as (3) economics and psychology. The impact of our research is evidenced by the high quality journal outlets, awards and media highlights. First, at the intersection of computer science and economics, we have made significant contributions. For example, IGERT fellow Elaine Wah and IGERT co-PI Michael Wellman studied high-frequency trading (HFT) published their findings in the ACM-EC'13. The bottom-line finding is that latency arbitrage (a particular form of HFT) reduces profits of non-HFT investors, and even degrades the efficiency of the market as a whole. The authors propose a discrete-time market clearing mechanism to eliminate latency arbitrage and improve market efficiency. This work was featured in Tech Crunch on June 14, 2013. (Wah and Wellman 2013) In Fall 2013, Elaine spearheaded a Michigan letter to the Commodity Futures Trading Commission (CFTC) outlining our position on stopping the latency arms race, which was co-signed by faculty in Finance and Law as well as Computer Science. Based on this, Wah and Wellman were invited to speak at CFTC, as well as the Chicago Federal Reserve. Elaine also had an op-ed in the Guardian a couple of weeks ago commenting on the Michael Lewis book. www.theguardian.com/commentisfree/2014/apr/04/michael-lewis-market-rigged-flash-boys-high-speed-trading The second interdisciplinary area our faculty and students have contributed to is at the intersection of computer science and political science. In U.S. politics, opinions on a variety of issues involving taxes, the role of government, domestic policy, and international relations are substantially though imperfectly correlated with each other and with party affiliation and with an overall self-identification as liberal or conservative. Thus, classifying people, media outlets, and opinions expressed in individual articles as liberal or conservative conveys meaning to most people. We applied three semi-supervised learning methods that propagate classifications of political news articles and users as conservative or liberal, based on the assumption that liberal users will vote for liberal articles more often, and similarly for conservative users and articles. We use data from the social news aggregator Digg, where readers vote for articles they think should be elevated to the front page. Starting from a few labeled articles and users, the algorithms propagate political leaning labels to the entire graph. In cross-validation, the best algorithm achieved 99.6% accuracy on held-out users and 96.3% accuracy on held-out articles. Adding social data such as users’ friendship or text features such as cosine similarity did not improve accuracy. The propagation algorithms, using the subjective liking data from users, also performed better than an SVM based text classifier, which achieved 92.0% accuracy on articles. The automatic classifier will be useful for a variety of scientific purposes, as well as in the development of news services that try to prevent political polarization by nudging people reading a mixture of both liberal and conservative items. (Zhou, Resnick and Mei 2011) The third interdisciplinary area is at the intersection of economics and psychology, also called behavioral economics. IGERT PI Yan Chen and her coauthors have published a series of papers investigating one of the most fundamental questions in economics, i.e., how to incorporate group structure in individual preferences. In the first of this series of papers, the PI and her co-author Sherry Li, present a laboratory experiment that measures the effects of induced group identity on participant social preferences. They find that participants are more altruistic towards an ingroup match. As a result, ingroup matching generates significantly higher expected earnings than outgroup matching. This paper combines experimental methods in economics and social psychology, and provides an empirical foundation for theoretical models of group identity. (Chen and Li, 2009) Applying this framework to an important problem domain, the PI and IGERT fellow Roy Chen study the effects of group identity on coordination. In this paper, they propose a group-contingent social preference model and derive conditions under which social identity changes equilibrium selection. They test their predictions in the minimum-effort game in the laboratory under parameter configurations which lead to an inefficient low-effort equilibrium for subjects with no group identity. For those with a salient group identity, consistent with their theory, they find that learning leads to ingroup coordination to the efficient high-effort equilibrium. Today's workplace is comprised of increasingly diverse social categories, including various racial, ethnic, religious and linguistic groups. This paper suggests that creating a unifying organization identity might be an effective mechanism in managing diversity. (Chen and Chen 2011)
