Sam Ganzfried

Email: sganz...@cs.cmu.edu

I am a PhD student in computer science at Carnegie Mellon, advised by Tuomas Sandholm, and I have an A.B. in math from Harvard. I am graduating in May 2015 and am seeking employment. Please contact me if you have information on an opportunity that you think would be a good fit. Here are my CV, Research Statement, and Teaching Statement.

My research interests are broadly in artificial intelligence and more specifically in computational game theory, multiagent systems, large-scale optimization, knowledge representation, data analysis and analytics, and multiagent learning. My PhD research has involved the design, theoretical analysis, and implementation of the most scalable approaches for computing strong game-theoretic strategies in imperfect-information games. Here is my thesis proposal, Computing Strong Game-Theoretic Strategies and Exploiting Suboptimal Opponents in Large Games.

An agent for two-player no-limit Texas Hold'em that I created (with Noam Brown and Tuomas Sandholm) came in first place in the 2014 Annual Computer Poker Competition. It beat each of the other submitted agents with statistical significance. Here are some articles about it: CMU, cardplayer part 1, cardplayer part 2. I describe the current state of research in computer poker in detail in this thread on the Two Plus Two forums.

I am organizing the AAAI-15 Workshop on Computer Poker and Imperfect Information, and I co-organized it in 2014 with Eric Jackson. I have been a PC member for AAAI (2012, 2014, 2015), AAMAS (2014), and IJCAI (2013, 2015).


Publications

Sam Ganzfried and Tuomas Sandholm. Safe Opponent Exploitation. ACM Transactions on Economics and Computation (TEAC), 2015. Special issue on selected papers from EC-12. Succeeds EC-12 paper listed below.

Sam Ganzfried and Tuomas Sandholm. Endgame Solving in Large Imperfect-Information Games. Manuscript, 2014. Succeeds AAAI-13 workshop paper listed below. Preliminary version in Proceedings of the Computer Poker and Imperfect Information Workshop at the AAAI Conference on Artificial Intelligence (AAAI), 2015.

Noam Brown, Sam Ganzfried, and Tuomas Sandholm. Hierarchical Abstraction, Distributed Equilibrium Computation, and Post-Processing, with Application to a Champion No-Limit Texas Hold'em Agent. Manuscript, 2014. Preliminary version in Proceedings of the Computer Poker and Imperfect Information Workshop at the AAAI Conference on Artificial Intelligence (AAAI), 2015. Short version appeared as Tartanian7: A Champion Two-Player No-Limit Texas Hold'em Poker-Playing Program in Proceedings of the Demonstrations Program at the AAAI Conference on Artificial Intelligence (AAAI), 2015.

Sam Ganzfried and Tuomas Sandholm. Potential-Aware Imperfect-Recall Abstraction with Earth Mover's Distance in Imperfect-Information Games. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2014.

Sam Ganzfried and Tuomas Sandholm. Action Translation in Extensive-Form Games with Large Action Spaces: Axioms, Paradoxes, and the Pseudo-Harmonic Mapping. In Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), 2013.

Sam Ganzfried and Tuomas Sandholm. Improving Performance in Imperfect-Information Games with Large State and Action Spaces by Solving Endgames. In Proceedings of the Computer Poker and Imperfect Information Workshop at the AAAI Conference on Artificial Intelligence (AAAI), 2013. Succeeded by 2014 manuscript listed above

Sam Ganzfried and Tuomas Sandholm. Tartanian5: A Heads-Up No-Limit Texas Hold'em Poker-Playing Program. In Proceedings of the Computer Poker Symposium at the AAAI Conference on Artificial Intelligence (AAAI), 2012.

Sam Ganzfried and Tuomas Sandholm. Safe Opponent Exploitation. In Proceedings of the ACM Conference on Electronic Commerce (EC), 2012. Succeeded by TEAC-15 paper listed above.

Sam Ganzfried, Tuomas Sandholm, and Kevin Waugh. Strategy Purification and Thresholding: Effective Non-Equilibrium Approaches for Playing Large Games. In Proceedings of the International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2012.

Sam Ganzfried. Computing Strong Game-Theoretic Strategies in Jotto. In Proceedings of the Conference on Advances in Computer Games (ACG), 2011.

Sam Ganzfried and Tuomas Sandholm. Game Theory-Based Opponent Modeling in Large Imperfect-Information Games. In Proceedings of the International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2011.

Sam Ganzfried and Tuomas Sandholm. Computing Equilibria by Incorporating Qualitative Models. In Proceedings of the International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2010. Extended version appeared as CMU technical report CMU-CS-10-105.

Sam Ganzfried and Tuomas Sandholm. Computing Equilibria in Multiplayer Stochastic Games of Imperfect Information. In Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), 2009.

Sam Ganzfried and Tuomas Sandholm. Computing an Approximate Jam/Fold Equilibrium for 3-Player No-Limit Texas Hold'em Tournaments. In Proceedings of the International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2008.

Sam Ganzfried. A New Algorithm for Knight's Tours. In Proceedings of the Research Experiences for Undergraduates Program in Mathematics, 2004.