Alex Strehl's Homepage

Alexander L. Strehl

Computer Science Ph.D. student
Rutgers University

Email: strehl@cs.rutgers.edu
Office(Lab): Hill 427
Office Phone: (732) 445-6430 Ext 6785
Office Phone (voicemail): (732) 445-6450 Ext 9504

Research Areas: Artificial Intelligence, Reinforcement Learning, Robotics, Combinatorics
Research Lab: Rutgers Laboratory for Real-Life Reinforcement Learning

I am an artificial-intelligence (AI) researcher at Rutgers University, in the Computer Science department. I am currently focused on solving problems in the exciting field of reinforcement learning and am extremely grateful to have Michael Littman as an advisor. My research has won several awards, including the Best Student Poster at the New York Academy of Sciences Machine-Learning Symposium in 2006 and the Best Paper Award at the International Conference on Tools with Artificial Intelligence (ICTAI) in 2004.

In 2002, I graduated from College of Charleston with a Bachelor of Science in Computer Information Systems and Mathematics. My undergraduate research, which involved the application of discrete mathematics to coding theory, a field of computer science, has also received numerous rewards. It was recognized by the Council on Undergraduate Research in their Mathematics and Computer Sciences Highlights and received first prize in the Mathematics category of the 2002 Scientific Research Poster Session at the College of Charleston. My undergraduate research advisor was Dinesh Sarvate.

Activities

I am defending my thesis on May 30th.

This semester (Spring 2007), I am helping to run the Rutgers Machine Learning Reading Group.

In the fall semester of 2006, I co-organized a light seminar, Learning Theory for Sequential Decision Making.

Publications

Working Journal Papers

Alexander L. Strehl, and Michael L. Littman. An Analysis of Model-Based Interval Estimation for Markov Decision Processes. Submitted in July 2006. pdf.

Reinforcement Learning

Alexander L. Strehl, Lihong Li, and Michael L. Littman. Incremental Model-based Learners With Formal Learning-Time Guarantees. In the proceedings of the 22nd Conference on Uncertainty in Artificial Intelligence (UAI 2006), pages 485-493, Cambridge, MA, USA, 2006. pdf.
Slides from my talk: ppt.
Alexander L. Strehl, Lihong Li, Eric Wiewiora, John Langford, and Michael L. Littman. PAC Model-Free Reinforcement Learning. In the proceedings of the 23rd International Conference on Machine Learning (ICML 2006), pages 881-888, Pittsburgh, PA, USA, 2006. pdf.
Longer Version with Appendix (Proofs of Lemmas 1 and 2): pdf.
Alexander L. Strehl, Chris Mesterharm, Michael L. Littman, and Haym Hirsh. Experience-Efficient Learning in Associative Bandit Problems. In the proceedings of the 23rd International Conference on Machine Learning (ICML 2006), pages 889-896, Pittsburgh, PA, USA, 2006. pdf.
Slides from my talk: ppt.
Carlos Diuk, Alexander L. Strehl, and Michael L. Littman. A Hierarchical Approach to Efficient Reinforcement Learning in Deterministic Domains. In the proceedings of the Fifth International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2006), Hakodate, Japan, 2006. pdf.
Alexander L. Strehl and Michael L. Littman. A Theoretical Analysis of Model-Based Interval Estimation. In the proceedings of the 22nd International Conference on Machine Learning (ICML 2005), pages 857-864, Bonn, Germany, 2005. pdf.
Longer Technical Report: pdf.
Bethany Leffler, Michael L. Littman, Alexander L. Strehl and Thomas Walsh, Efficient Exploration with Latent Structure In the proceedings of Robotics: Science and Systems (RSS 2005), Cambridge, MA, USA, 2005. pdf.
Alexander L. Strehl and Michael L. Littman. An Empirical Evaluation of Interval Estimation for Markov Decision Processes. In the proceedings of the 16th IEEE International on Tools with Artificial Intelligence Conference (ICTAI 2004), pages 128-135, Boca Raton, FL, USA, 2004. pdf.

Combinatorics

Alexander L. Strehl. Ternary Codes Through Ternary Designs. In the Australasian Journal of Combinatorics (AJC), 30, 21-30, 2004. pdf.
Dinesh G. Sarvate and Alexander L. Strehl. Linear Codes Through Latin Squares. In the Bulletin of the Institute of Combinatorics and Its Applications, 27, 73-81, 2003. pdf.

Workshop Papers

Alexander L. Strehl, Lihong Li, and Michael L. Littman. PAC Reinforcement Learning Bounds for RTDP and Rand-RTDP. To be presented at the Twenty-First National Conference on Artificial Intelligence (AAAI 2006) Workshop on Learning For Search, Boston, MA, USA, 2006. pdf.
Longer Technical Report: pdf.
Slides from my talk: ppt.
Carlos Diuk, Michael L. Littman, and Alexander L. Strehl. A Hierarchical Approach to Efficient Reinforcement Learning in Factored State Spaces. Presented at the the 22nd International Conference on Machine Learning (ICML 2005), Workshop on Rich Representations for Reinforcement Learning, Bonn, Germany, 2005.

Theses

Alexander L. Strehl. N-ary Codes, Designs, and Enclosings. Bachelor's Essay to satisfy my undergraduate degree requirements for the Honors College at the College of Charleston, 2002. pdf.