Brian Milch

Multi-agent influence diagrams for representing and solving games

The traditional representations of games using the extensive form or the strategic (normal) form obscure much of the structure that is present in real-world games. In this paper, we propose a new representation language for general multiplayer games--multi-agent influence diagrams (MAIDs). This representation extends graphical models for probability distributions to a multi-agent decision-making context. MAIDs explicitly encode structure involving the dependence relationships among variables. As a consequence, we can define a notion of strategic relevance of one decision variable to another: D′ is strategically relevant to D if, to optimize the decision rule at D, the decision maker needs to take into consideration the decision rule at D′. We provide a sound and complete graphical criterion for determining strategic relevance. We then show how strategic relevance can be used to detect structure in games, allowing a large game to be broken up into a set of interacting smaller games, which can be solved in sequence. We show that this decomposition can lead to substantial savings in the computational cost of finding Nash equilibria in these games.

Query-free news search

Many daily activities present information in the form of a stream of text, and often people can benefit from additional information on the topic discussed. TV broadcast news can be treated as one such stream of text; in this paper we discuss finding news articles on the web that are relevant to news currently being broadcast.We evaluated a variety of algorithms for this problem, looking at the impact of inverse document frequency, stemming, compounds, history, and query length on the relevance and coverage of news articles returned in real time during a broadcast. We also evaluated several postprocessing techniques for improving the precision, including reranking using additional terms, reranking by document similarity, and filtering on document similarity. For the best algorithm, 84%-91% of the articles found were relevant, with at least 64% of the articles being on the exact topic of the broadcast. In addition, a relevant article was found for at least 70% of the topics.

Searching the Web by voice

Spoken queries are a natural medium for searching the Web in settings where typing on a keyboard is not practical. This paper describes a speech interface to the Google search engine. We present experiments with various statistical language models, concluding that a unigram model with collocations provides the best combination of broad coverage, predictive power, and real-time performance. We also report accuracy results of the prototype system.

First-Order Probabilistic Languages: Into the Unknown

This paper surveys first-order probabilistic languages(FOPLs), which combine the expressive power of first-order logic with a probabilistic treatment of uncertainty. We provide a taxonomy that helps make sense of the profusion of FOPLs that have been proposed over the past fifteen years. We also emphasize the importance of representing uncertainty not just about the attributes and relations of a fixed set of objects, but also about what objects exist. This leads us to Bayesian logic, or BLOG, a language for defining probabilistic models with unknown objects. We give a brief overview of BLOG syntax and semantics, and emphasize some of the design decisions that distinguish it from other languages. Finally, we consider the challenge of constructing FOPL models automatically from data.

AI's 10 to Watch

The recipients of the 2008 IEEE Intelligent Systems 10 to Watch award—Philipp Cimiano, Dmitri Dolgov, Anat Levin, Peter Mika, Brian Milch, Louis-Philippe Morency, Boris Motik, Jennifer Neville, Erik Sudderth, and Luis von Ahn—discuss their current research and their visions of AI for the future.