Aaron Arvey Mission Work Experience Motivated student with research experience in computer science and mathematics. Excellent work ethic, seeking challenging research position. Department of Computer Science University of California at San Diego 9500 Gilman Drive Mail Code 0404 La Jolla, CA 92093 Toyon Research – Mark Meloon Email: aarvey@cs.ucsd.edu Phone: (858) 534-5617 Santa Barbara, CA Summer 2005 Research Intern – Algorithm Development Researched methods for inference in temporal domains and implemented (C++) the Boyen-Koller algorithm to perform approximate online inference for hierarchical dynamic Bayesian networks. Implemented system integration using dynamic Bayesian networks to infer intent of a group. This is also known as level 2 fusion within the context of the Joint Directorate Laboratories (JDL). Reed Inst. & MARC Math – Janet Myhre Claremont, CA 2004-2006 Senior Research Fellow & Consultant – Computational Statistics Performed research into numerical methods for statistical computing, specifically Bayesian data adjustment, and contractive map methods. Implemented numerical algorithm for Lockheed Martin and the Navy which computes statistical lower bounds of reliability on complex (e.g. non-series) systems, such as missiles. Sandia National Labs – Steve Hurd Livermore, CA Summer 2004 Research Intern Researched methods for unsupervised machine learning – specifically statistical clustering algorithms for outlier detection. Performed red team exercise to enhance security practices. Programmed IXA2400 network card for accelerated specialized processing of network packets. This ability was then used to perform outlier detection. SOS Mellon Grant – Janice Marino Claremont, CA Fall 2002-Present Database and Web Programmer Configured Linux server, installed database operations, and migrated data from old infrastructure. Created MySQL databases and PHP-driven web user-interfaces relating to job monitoring, productivity, workshop registration, inventory, and finances. Worked with another programmer under a manager as a team to design the databases and front ends. Research University of California – Yoav Freund Assistantships Machine Learning San Diego, CA 2006-2007 Developed new boosting algorithm based on non-convex loss. Implemented the new boosting algorithm and created a SourceForge open source repository. Performed several applications of new boosting algorithm. Keck Graduate Institute – Alpan Raval Claremont, CA 2005-2006 Bioinformatics and Computational Biology Extended results of previous study to include higher order Markov assumptions for the Jensen Shannon divergence method. Implemented Monte Carlo simulation across a commodity cluster of computers to determine significance of coding signal in DNA based on higher order Jensen Shannon divergence. Implemented recursive segmentation algorithms to detect coding vs. non-coding regions in DNA. Reed Institute – Janet Myhre Claremont, CA 2005-2006 Senior Research Fellow - Statistical Algorithm Development Performed statistical analysis of reliability data using multiple measures such as fault trees, approximation methods using empirical Bayesian data adjustment, and classical frequentist methods. Assisted in designing efficient numerical algorithms for analyzing Trident II missile reliability lower confidence bounds. Mathematics Dept – Asuman Aksoy Claremont, CA 2005-2006 Problem and Solution Editor Teaching Experience Professor Aksoy is writing a book for undergraduates taking an introductory analysis class. I am helping develop the problem and solution portions. Responsibilities include first-round editing and typesetting the book using LaTeX. Harvey Mudd College – Belinda Thom Claremont, CA Spring 2005 Machine Learning Helped teach an upper division course in Machine Learning. Assisted in lecture ideas, designed homework describing algorithms and implementations. Presented lectures on information theory and its applications in machine learning. Claremont McKenna College - Greg Stein Claremont, CA 2004-2005 “Grutor” for Discrete Math Graded homework and tutored students Assisted in determining questions for exams and homework problems. Self Supervised Seattle, WA 2001-2002 Teacher of Computer Skills in Spanish Education Created a course intended to teach basic computer skills. Taught a diverse group from South and Central America in colloquial Spanish. Communicated on different levels to accommodate different learning speeds and styles. University of California San Diego, CA Graduating 2006 PhD in Computer Science Publishing work in bioinformatics and boosting. Claremont McKenna College Claremont, CA Graduated 2006 BA Degree in Computer Science and BA Degree in Mathematics Research interests include computational biology, probabilistic inference, and general machine learning. Double major in mathematics at Claremont McKenna and computer science at Harvey Mudd. North Seattle Community College Seattle, WA Graduated 2002 Associate of Arts Degree Awards Vice President of the Associated Student Body NSF Honorable Mention Research grant and fellowship from Reed Institute for Decision Science to research computational methods in reliability theory. Presenting results to Lockheed Martin. Awarded the Janet Myhre Merit Scholarship for excellence in applied statistics Travel grants from Claremont McKenna College Publications Full papers: o Ron Alterovitz, Sriram Sankararaman, Aaron Arvey, Yoav Freund, and Kimmen Sjolander. Predicting Catalytic Residues Using Protein Sequence and Structure Features. Journal submission in progress. o Aaron Arvey, Alpan Raval, Rajeev Azad, and Jeffrey Lawrence. Generalized JensenShannon divergence measure for sequence analysis. Submitted to The Journal of Theoretical Biology. o Aaron Arvey and Eric Aaron. Online Markov Decision Processes For Learning Movement. Proceedings of the 6th International Conference on Computer Games: Artificial Intelligence and Mobile Simulation, pg 48-53, 2005. Technical Reports o Mike Reine, Aaron.Arvey, Janet Myhre. DFR Lower Confidence Bounds for LE Subpopulations. Submitted to Lockheed Martin. November 2005. o Aaron Arvey. Hierarchical Dynamic Bayesian Networks for Level 3 Fusion. Submitted to Toyon Research Corporation. September 2005. o Aaron Arvey, Cal Pierog, Ed Kim, Evan Parry, Elizabeth Sweedyk, and Dan Kegel. Differential Code Coverage in the Context of the Wine Project. Submitted to Google and Computer Science Department of Harvey Mudd College. May 2005. Posters (Abstracts): o Aaron Arvey. Markovian Change Point Analysis to Detect Coding Regions in DNA. Joint MAA and AMS Meeting Undergraduate Research Poster Session. San Antonio, Texas. Janurary 2006. (Excellent Poster Award) o Aaron Arvey. Empirical Bayesian Data Adjustment and Subpopulations. Southern California Meeting of the American Statistics Association. November 2005. (Best Undergraduate Poster) o Aaron Arvey and David Nichols. Asymptotically Periodic Solutions of Linear ODE’s. Joint MAA and AMS Meeting Undergraduate Research Poster Session. Phoenix, Arizona. Janurary 2004. (Excellent Poster Award). Also presented extensions at other poster sessions.Posters (Abstracts): Computer Skills Fluent in C, C++, PHP, Python, Java, and MySQL Able to learn new languages and concepts extremely quickly Comfortable with Perl, SML, Bash, Visual Studio IDE, and low level techniques in assembly (x86/mips) Able to contribute and create open source projects. Skills Excellent communication Entrepreneurship and initiative Customer service Ability to adapt and adjust to situational needs Quick and independent learning Integrate mathematics and computer science