Christopher Lewis DuBois

Contact Information University of California, Irvine Department of Statistics School of Information and Computer Science 3019 Bren Hall Irvine, CA 92617-5100 USA University of California, Irvine, CA Department of Statistics - Graduate Student Pomona College, Claremont, CA B.A., Applied Mathematics Minor in Physics Jesuit High School Portland, OR Diploma Research Interests [email protected]

fax: +1 949.824.4056 http://www.ics.uci.edu September 2008 - Current

Education

May 2006

June 2002

Statistical models for relational events, e.g. communication or interaction within a social network. In large-scale settings, models which summarize the network (e.g. latent space models) may be able to share statistical strength across less observed portions of the network in order to increase predictive accuracy. I would like to investigate these models in a variety of online contexts (e.g. email, question and answer forums, etc). Christopher DuBois, James Foulds, Padhraic Smyth. Latent set models for co-appearance data. Fifth International AAAI Conference on Weblogs and Social Media (ICWSM), 2011. James Foulds, Arthur U. Asuncion, Christopher DuBois, Carter T. Butts, Padhraic Smyth. Fourteenth International Conference on Articial Intelligence and Statistics (AISTATS), 2011. Christopher DuBois, Padhraic Smyth. Modeling Relational Events via Latent Classes. Proceedings of the 12th Conference on Knowledge Discover and Data Mining (KDD), 2010. Perry G. Schiro, Christopher L. DuBois, and Alfred S. Kwok. Large capture-range of a single-beam gradient optical trap. Optics Express, 11 (25) p.3485. 2003.

Publications

Invited Talks

Bayesian Latent Set Models for Co-appearance Data. Institute for Mathematical Behavioral Sciences: Behavior In and On Networks. January 21-23, 2011 Christopher DuBois, Daniel McFarland, Padhraic Smyth, Carter Butts. Hierarchical Models of High School Classroom Dynamics. Workshops on Information Networks, 2011. New York, NY. Christopher DuBois, Padhraic Smyth. Bayesian set models for two-mode data. Sunbelt Social Networks Conference, 2011. Clearwater, FL. Christopher DuBois, Padhraic Smyth. Modeling Relational Events via Latent Classes. Proceedings of the 12th Conference on Knowledge Discover and Data Mining (KDD), 2010. Washington DC. Christopher DuBois, Padhraic Smyth. Modeling Relational Events via Latent Classes. Sunbelt Social Networks Conference, 2010. Italy.

Conference Presentations

Poster Presentations

Christopher DuBois, James Foulds, Padhraic Smyth. Latent set models for co-appearance data. Southern California Machine Learning Workshop (SoCaML), 2011. Christopher DuBois, James Foulds, Padhraic Smyth. Latent set models for co-appearance data. Fifth International AAAI Conference on Weblogs and Social Media (ICWSM), 2011. Christopher DuBois, Carter Butts, Padhraic Smyth. Testing the Theory of Structural Holes: Modeling the Microbehavior of an Email Network. Sunbelt Social Networks Conference, 2009. San Diego, CA.

Grants, Fellowships, Awards

National Defense Science and Engineering Graduate Fellowship United States Department of Defense Chairs Fellowship Department of Statistics, University of California, Irvine, CA Howard Hughes Medical Institute Undergraduate Summer Research Grant Pomona College, Claremont, CA Pomona College Summer Undergraduate Research Program Grant Pomona College, Claremont, CA Pomona College Scholar, Claremont, CA

2009-2012

2008-2009

2005

2003, 2004

Spring, 2005

Professional Experience

Google Research, New York, NY Decision Support Engineering Analyst June 2009 - September 2009 Worked with Daryl Pregibon on advertiser coappearance graphs. Presented ndings in a presentation for Google Research as well as a poster conference for Google interns. HealthOne Stang, Portland, OR Director of Nurse Networking / Web Developer April, 2005 - April, 2006 I executed a six-month project integrating a hosted database solution with the companys new website, working closely with web designers, developers, upper management, nurses, and the team of recruiters to ensure that my development eorts satised all usability requirements. CafeMax Cycling Team, Alcoy, Spain Domestique January, 2007 - August, 2007 Raced for an amateur road cycling team, I competed in races all over Spain, including several 4-day tours support of the team leader. Strategic Solutions NW, Portland, OR Consultant June 2006 - January 2007 Helped transition a paper company to new accounting software after they acquired six mills. Participated in accounts payable and software conguration tasks. ShareThis, Inc, Portland, OR Usability Engineer Intern June, 2001 - August, 2001 Tested user interface for a Java web client instant messaging service.

Research Experience

Department of Computer Science, University of California - Irvine, CA Graduate Student Researcher August 2008 - Current Researching models for massive networks of online collaboration and communication (e.g. Wikipedia, email). We are studying latent space models and evaluating their predictive accuracy on unobserved events. University of Otago - Dunedin, New Zealand Conference Participant

December 2007

In the winter of 2007 I participated in the Workshop on Applications of Mathematics to Biomedicine at the University of Otago in Dunedin, New Zealand. I worked with a small, multidisciplinary team to solve problems in pharmaceutical research, such as predicting drug release kinetics of common tablets. This work may be published as a working paper in coming months. Department of Mathematics, Pomona College - Claremont, CA Student Researcher June 2004 - August 2005 Tumors initiate the growth of blood vessels that are more tortuous, clustered, and irregular than vasculature found in normal tissue. In the spring of 2006, I employed techniques from computational topology to develop a protocol that would distinguish between scanning electron microscope images of tumor vessels and normal vessels. I presented this work in my undergraduate thesis in applied mathematics.

Under a grant from Howard Hughes Medical Institute, I helped develop a cellular automata model for melanoma growth. The model correctly showed necrotic core development while considering only microenvironmental conditions (e.g. pH, glucose concentrations, and so on). We presented this work to a group of oncologists at UCLA. Department of Physics, Pomona College - Claremont, CA Student Researcher

June 2003 - August 2003

Fellow student Perry Schiro and I investigated the long-range capabilities of laser tweezers, the technique of trapping small particles approximately 10 microns in diameter in the focal point of an infrared laser beam. I calibrated equipment, designed experiments, and performed data analysis. I presented this work in a peer-reviewed article and at an undergraduate research symposium. Competitions R Package Recommendation Engine: 8th place (of 57 teams). February 2011. (kaggle.com) Hearst Data Mining Challenge: 10th place (of 61 teams). November 2010. (hearstchallenge.com via kaggle.com American Statistical Association (October 2008 - Present) Sigma Xi (June 2005 - Present) International Network for Social Network Analysis (INSNA) (December 2008 - Present) Teaching Assistant. Math and Music, with Professor Ami Radunskaya. Spring, 2004. Dr. Padhraic Smyth, Associate Professor University of California, Irvine Department of Computer Science 3019 Bren Hall, Irvine, CA 92697-3435 [email protected] Dr. Ami Radunskaya, Associate Professor Pomona College Depatrment of Mathematics 610 North College Avenue, Claremont, CA 91711 [email protected] Dr. Daryl Pregibon Google Research 76 9th AVE, 4th oor, New York, NY 10011 [email protected]

Professional Memberships

Teaching Experience References