Gavin Taylor
Ph.D. Student, Computer Science
Duke University
I am working with Dr. Ron Parr on several projects relating to linear approximations of MDP models and value functions. My Research Initiation Project focuses on uncovering links between model approximations and value function approximations. Besides Ron Parr, some of this work is done with Christopher Painter-Wakefield, and Dr. Michael Littman and Lihong Li of Rutgers.
Besides Dr. Parr’s research group, I participate in the Duke Robotics, Intelligence, and Vision (DRIV) reading group as well. For more information about Duke Artificial Intelligence and Vision research, please see the DRIV Wiki linked to on the bar below.
Research
Graduate Coursework
Fall 2006
CPS 230 - Algorithms
CPS 270 - Artificial Intelligence
CPS 258 - Scientific Computing
Spring 2007
CPS 210 - Operating Systems
CPS 296 - Robotics
Fall 2007
CPS 271 - Machine Learning
CPS 229 - Computer Architecture
Spring 2008
STA 244 - Linear Models
Contact Information
E-mail: gvtaylor (AT) cs (DOT) duke (DOT) edu
Office: North 207
Office Phone: 919/660-4020
Spring 2007
CPS 100
CPS 100E
Fall 2007
CPS 100
Refereed Conference Papers
Ronal Parr, Lihong Li, Gavin Taylor, Christopher Painter-Wakefield, and Michael Littman. An Analysis of Linear Models, Linear Value-Function Approximation, and Feature Selection for Reinforcement Learning. To appear in ICML 2008.
Published Undergraduate Work
Campbell, A. and L.J. Heyer, et al. Microarrays for the Masses: Pedagogical Resources for High School through College. Abstract for the American Society of Cell Biology. December, 2007.
Teaching
In the summer of 2008, Mac Mason and I are designing and teaching a brand new course on Robotics for two sessions of the Duke TIP program. The course website can be found here. Additionally, I have held two semesters of Teaching Assistantships.