Aaron Arvey Box #142
Aaron Arvey is currently an undergraduate student at Claremont McKenna College. He is completing majors in Computer Science (at Harvey Mudd College) and Mathematics (at McKenna). His research interests include probabilistic inference in machine learning, bioinformatics, and neural processing sytems. He is a senior fellow of the Reed Institute for Decision Science, Myhre scholar, consultant for the U.S. Navy and Lockheed Martin at MARC Consulting Group, and research assistant at Keck Graduate Institute in computational biology.
Online Markov Decision Processes for Learning Movement in Games. Proceedings of the 6th International Conference on Computer Games: Artificial Intelligence and Mobile Systems. July 2005. Paper and videos
(In Preparation) Higher Order Jensen-Shannon Segmentation of Symbolic Sequences. Physical Review E.
Asymptotically Periodic Solutions of Linear ODE's Joint MAA and AMS Meeting Undergraduate Research Poster Session. Phoenix, Arizona. January 2004. Abstract, poster, and presentation (Outstanding Poster Award)
Empirical Bayesian Data Adjustment and Subpopulations Southern California Meeting of the American Statistics Association. November 2005. presentation (Best Undergraduate Poster)
Markovian Change Point Analysis to Detect Coding Regions of DNA Joint MAA and AMS Meeting Undergraduate Research Poster Session. San Antonio, Texas. January 2006. presentation (Outstanding Poster Award)
Hierarchical Dynamic Bayesian Networks and Level 3 Fusion. Submitted to Toyon Research Corporation. September 2005.
DFR Lower Confidence Bounds and LE Subpopulations. Submitted to Lockheed Martin and the Navy. November 2005.
I have given several presentations on topics such as
I'll add links to the actual presentations soon!
I have taken classes in computer science and math. While my main interest is in Machine Learning, I have never taken any formal course in the subject matter.
Reed Institute for Decision Science sponsored research in the field of statistical analysis of reliability data using multiple measures such as fault trees, approximation methods using Bayesian data adjustment, and classical frequentist methods.
Keck Graduate Institute (KGI) funded work for Monte Carlo simulation of biological symbolic data for recursive segmentation algorithms as applied to the problem of detecting coding vs. non-coding regions.
In the 2005 spring semester, I assisted Professor Belinda Thom teach a upper division course in machine learning. I designed and wrote (along with keys) several homework and lab assignments. Seeing as how I was a junior and all the students in the course were seniors, I asked to not grade any of the assignments
I graded for a discrete math course taught by Professor Greg Stein at Claremont McKenna College where my only responsibility was to correct assignments and tutor students.
While at community college, I used my student government office (surprisingly large) to teach a group of Spanish speaking students the basic computer skills necessary to succeed in a college environment. At the time I spoke Spanish fluently. Skills taught included basic operations in MS Windows, MS Office, and internet communication.
I am a member of the mathematical societies AMS and MAA. I am a member of AUAI (Association for Uncertainty in Artificial Intelligence). I'm soon to be a member of the ACM and IEEE.
Why am I a member? Well, for the AMS and MAA, I received free membership due to my research in areas of interest to these groups. My main research is in uncertainty reasoning, and the mailing lists for AUAI also provide insight into conferences that I would be otherwise unaware of. I'm going to join IEEE because there are several articles I would like to read which are only accessible to members. The ACM special interest groups are an excellent resource to gain insight in to what other people in similar research areas are currently persuing. While I was putting together an article for Google, I wanted to find all I could on software engineering and the ACM special interest group provided me all the resources I could ask for... even better then "googling" for software engineering!