Kevin Waugh

About Me

February 23, 2010
E-mail:my last name–at–cs.cmu.edu
Position:PhD Student
Advisor:Drew Bagnell
Office:GHC 9223
Phone:412-268-1405
Fax:412-268-5576
Mail: Carnegie Mellon University
Department of Computer Science
5000 Forbes Avenue
Pittsburgh, PA 15213-3891
Education: PhD in Computer Science, Carnegie Mellon University, 2009 to present.
MSc in Computing Science, University of Alberta, 2007 to 2009.
BSc in Computing Science, University of Alberta, 2002 to 2007.

Research Interests

August 30, 2013

I am interested in Artificial Intelligence, Machine Learning, and Game Theory. In particular, I find convex optimization, online learning and how the two relate fascinating. My thesis work focuses on predicting purposeful behavior in strategic situations from only a few observations. In our paper, Computational Rationalization: The Inverse Equilibrium Problem, we predict the location of new mid-scale hotels at a county level using demographic and regulatory features.

My master's thesis work at the University of Alberta, under the supervision of Michael Bowling and Dale Schuurmans, involved Computer Poker programs and Computational Game Theory. You can learn about their on going work at the Computer Poker Research Group website.

Publications

December 2, 2014

2015

Solving Games with Functional Regret Estimation. Kevin Waugh, Dustin Morrill, J. Andrew Bagnell, and Michael Bowling.In AAAI Conference on Artificial Intelligence, 2015. BibTeX     Download: [pdf] [arXiv] 
A Unified View of Large-scale Zero-sum Equilibrium Computation. Kevin Waugh and J. Andrew Bagnell.In AAAI Workshop on Computer Poker and Imperfect Information, 2015. BibTeX     Download: [pdf] [arXiv] 

2013

A Fast and Optimal Hand Isomorphism Algorithm. Kevin Waugh.In AAAI Workshop on Computer Poker and Imperfect Information, 2013. BibTeX     Download: [pdf] 
Computational Rationalization: The Inverse Equilibrium Problem. Kevin Waugh, Brian D. Ziebart, and J. Andrew Bagnell.Technical Report arXiv 1308.3506, 2013. BibTeX     Download: [arXiv] 

2011

Accelerating Best Response Calculation in Large Extensive Games. Michael Johanson, Kevin Waugh, Michael Bowling, and Martin Zinkevich.In Proceedings of the Twenty-Second International Joint Conference on Artificial Intelligence (IJCAI), 2011. BibTeX     Download: [pdf] 
Computational Rationalization: The Inverse Equilibrium Problem. Kevin Waugh, Brian D. Ziebart, and J. Andrew Bagnell.In Proceedings of the 28th International Conference on Machine Learning (ICML), 2011. Best paper. BibTeX     Download: [pdf] [arXiv] 

2009

Monte Carlo Sampling for Regret Minimization in Extensive Games. Marc Lanctot, Kevin Waugh, Martin Zinkevich, and Michael Bowling.In Advances in Neural Information Processing Systems 22 (NIPS), pp. 1078–1086, 2009. A longer version is available as a University of Alberta Technical Report, TR09-15. BibTeX     Download: [pdf] 
Strategy Grafting in Extensive Games. Kevin Waugh, Nolan Bard, and Michael Bowling.In Advances in Neural Information Processing Systems 22 (NIPS), pp. 2026–2034, 2009. BibTeX     Download: [pdf] 
Abstraction in Large Extensive Games. Kevin Waugh.Master's Thesis, University of Alberta,2009. BibTeX     Download: [pdf] 
Faster State Manipulation in General Games using Generated Code. Kevin Waugh.In Proceedings of the 1st General Intelligence in Game-Playing Agents (GIGA), 2009. BibTeX     Download: [pdf] 
A Practical Use of Imperfect Recall. Kevin Waugh, Martin Zinkevich, Michael Johanson, Morgan Kan, David Schnizlein, and Michael Bowling.In Proceedings of the 8th Symposium on Abstraction, Reformulation and Approximation (SARA), 2009. To appear. BibTeX     Download: [pdf] 
Abstraction Pathologies in Extensive Games. Kevin Waugh, David Schnizlein, Michael Bowling, and Duane Szafron.In Proceedings of the 8th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), pp. 781–788, 2009. BibTeX     Download: [pdf] 

2007

Interactive Story Authoring: A Viable Form of Creative Expression for the Classroom. Mike Carbonaro, Maria Cutumisu, Harvey Duff, Stephanie Gillis, Curtis Onuczko, Jeff Siegel, Jonathan Schaeffer, Allan Schumacher, Duane Szafron, and Kevin Waugh. Computers and Education, 51(2):687–707, 2007. BibTeX     Download: [pdf] 
ScriptEase: A Generative/adaptive Programming Paradigm for Game Scripting. Maria Cutumisu, Curtis Onuczko, Mattew McNaughton, Thomas Roy, Jonathan Schaeffer, Allan Schumacher, Jeff Siegel, Duane Szafron, Kevin Waugh, Mike Carbonaro, Harvey Duff, and Stephanie Gillis. Science of Computer Programming, 67(1):32–55, 2007. BibTeX     Download: [pdf] 

2006

Adapting a Commercial Role-Playing Game for Educational Computer Game Production. Mike Carbonaro, Maria Cutumisu, Harvey Duff, Stephanie Gillis, Curtis Onuczko, Jonathan Schaeffer, Allan Schumacher, Jeff Siegel, Duane Szafron, and Kevin Waugh.In Proceedings of GameOn 06, pp. 54–61, 2006. BibTeX     Download: [pdf] 

2005

A Pattern Catalog for Computer Role Playing Games. Curtis Onuczko, Maria Cutumisu, Duane Szafron, Jonathan Schaeffer, Matthew McNaughton, Thomas Roy, Kevin Waugh, Mike Carbonaro, and Jeff Siegel.In Proceedings of GameOn 05, pp. 33–38, 2005. BibTeX     Download: [pdf] 
Generated by bib2html.pl written by Patrick Riley.

Code and Downloads

August 30, 2013

Here are some programs I have written that you may find interesting or useful.

Hand Isomorphism

A fast hand indexing library written in C. In poker, a card's suit is of no strategic importance, and the order in which the cards are dealt in a round is also irrelevant. This library provides a fast mapping between a poker hand to a canonical index, which is the same for all isomorphic hands.

Matrix game ICE

This python software predicts the behavior of the players in a matrix game using the Inverse Correlated Equilibrium technique described in our ICML paper, Computational Rationalization: The Inverse Equilibrium Problem.

OpenCFR

An implementation of the chance-sampled counterfactual regret minimization algorithm for Leduc Hold'em. It allows for one to solve various abstract games of Leduc Hold'em and compute the exploitability of the resulting strategies.

rank7

A fast seven card poker hand rank evaluator. It uses a series of small lookup tables to determine the rank of a poker hand.

zerosum

A zero-sum extensive-form game package.

You can find my latest open source software on my GitHub page.

Interests and Hobbies

September 25, 2009

In my spare time I like to:

Kevin Waugh