Being able to compactly represent large state spaces is crucial in solving a vast majority of practical stochastic planning problems. This requirement is even more stringent in the context of multi-agent systems, in which the world to be modeled also includes the mental state of other agents. This leads to a hierarchy of beliefs that results in a continuous, unbounded set of possible interactive states, as in the case of Interactive POMDPs. In this paper, we describe a novel representation for interactive belief hierarchies that combines first-order logic and probability. The semantics of this new formalism is based on recursively partitioning the belief space at each level of the hierarchy; in particular, the partitions of the belief simplex at one level constitute the vertices of the simplex at the next higher level. Since in general a set of probabilistic statements only partially specifies a probability distribution over the space of interest, we adopt the maximum entropy principle in order to convert it to a full specification.

Conceptually founded on the "patternist" systems theory of intelligence outlined in (Goertzel 2006), OCP combines Defining Forgetting In ordinary human discourse, the multiple AI paradigms such as uncertain logic, computational word "forget" has multiple shades of meaning. It can refer linguistics, evolutionary program learning and connectionist to the irreversible elimination of a certain knowledge item attention allocation in a unified architecture. Cognitive from memory; or it can mean something milder, as in cases processes embodying these different paradigms interoperate where someone "forgets" something, but then remembers it together on a common neural-symbolic knowledge shortly after. In the latter case, "forgetting" means that the store called the Atomspace. The interaction of these processes knowledge item has been stored in some portion of memory is designed to encourage the self-organizing emergence from which access is slow and uncertain.

Autonomous robots demand complex behavior to deal with unstructured environments. To meet these expectations, a robot needs to address a suite of problems associated with long term knowledge acquisition, representation, and execution in the presence of partial information. In this paper, we address these issues by the acquisition of broad, domain general skills using an intrinsically motivated reward function. We show how these skills can be represented compactly and used hierarchically to obtain complex manipulation skills. We further present a Bayesian model using the learned skills to model objects in the world, in terms of the actions they afford. We argue that our knowledge representation allows a robot to both predict the dynamics of objects in the world as well as recognize them.

The social world that most of us navigate effortlessly can prove to be a perplexing and disconcerting place for individuals with autism. Currently there are no models to assist non-expert authors as they create customized social script-based instructional modules for a particular child. We describe an approach to using human computation to develop complex models of social scripts for a plethora of complex and interesting social scenarios, possible obstacles that may arise in those scenarios, and potential solutions to those obstacles. Human input is the natural way to build these models, and in so doing create valuable assistance for those trying to navigate the intricacies of a social life.

Analyzing chat traffic has important applications for both the military and the civilian world. This paper presents a case study of a real-world application of chat analysis in support of team training exercise in the military. It compares the results of an unsupervised learning approach with those of a supervised classification approach. The paper also discusses some of the specific challenges presented by this domain.

Consider an embedded agent with a self-modifying, Turing-equivalent policy that can change only through active self-modifications. How can we make sure that it learns to continually accelerate reward intake? Throughout its life the agent remains ready to undo any self-modification generated during any earlier point of its life, provided the reward per time since then has not increased, thus enforcing a lifelong success-story of self-modifications, each followed by long-term reward acceleration up to the present time. The stack-based method for enforcing this is called the success-story algorithm. It fully takes into account that early self-modifications set the stage for later ones (learning a learning algorithm), and automatically learns to extend self-evaluations until the collected reward statistics are reliable ... a very simple but general method waiting to be re-discovered! Time permitting, I will also briefly discuss more recent mathematically optimal universal maximizers of lifelong reward, in particular, the fully self-referential Goedel machine.

Britain's trains are not noted for their AAAI Activity Context Representation Workshop. The punctuality and they are deemed on-time within a window first paper, 'Defining and Representing Activity Context of ten or so minutes, so just using the train timetable to for Systems Analysis', summarizes the author's formal predict bad spots is not feasible. Over a number of journeys, Simplified Set Theory (SST) approach and the use of his the user attempts to find journey landmarks that precede PentaVenn diagram. This second paper uses these in a the bad spots by a few minutes ("a few" being less modest, partially worked example to explore the contexts than the predicted time for file transfer). Some landmarks of an activity and how a formal approach can aid systems might be easy to identify, e.g.

Discovering latent connections between seemingly unconnected documents and constructing "stories" from scattered pieces of evidence are staple tasks in intelligence analysis. We have worked with government intelligence analysts to understand the strategies they use to make connections. Beyond techniques like clustering that aim to provide an initial broad summary of large document collections, an important goal of analysts in this domain is to assimilate and synthesize fine grained information from a smaller set of foraged documents. Further, analysts' domain expertise is crucial because it provides rich contextual background for making connections and thus the goal of KDD is to augment human discovery capabilities, not supplant it. We describe a visual analytics system we have built - Analyst's Workspace (AW) - that integrates browsing tools with a storytelling algorithm in a large screen display environment. AW helps analysts systematically construct stories of desired fidelity from document collections and helps marshall evidence as longer stories are constructed.

Frequently, it is advantageous for an agent to model other agents in order to predict their behavior during an interaction. Modeling others as rational has a long tradition in AI and game theory, but modeling other agents’ departures from rationality is difficult and controversial. This paper proposes that bounded rationality be modeled as errors the agent being modeled is making while deciding on its action. We are motivated by the work on quantal response equilibria in behavioral game theory which uses Nash equilibria as the solution concept. In contrast, we use decision-theoretic maximization of expected utility. Quantal response assumes that a decision maker is rational, i.e., is maximizing his expected utility, but only approximately so, with an error rate characterized by a single error parameter. Another agent’s error rate may be unknown and needs to be estimated during an interaction. We show that the error rate of the quantal response can be estimated using Bayesian update of a suitable conjugate prior, and that it has a finitely dimensional sufficient statistic under strong simplifying assumptions. However, if the simplifying assumptions are relaxed, the quantal response does not admit a finite sufficient statistic and a more complex update is needed. This confirms the difficulty of using simple models of bounded rationality in general settings.

There has been significant recent interest in computing effective practical strategies for playing large games. Most prior work involves computing an approximate equilibrium strategy in a smaller abstract game, then playing this strategy in the full game. In this paper, we present a modification of this approach that works by constructing a deterministic strategy in the full game from the solution to the abstract game; we refer to this procedure as purification. We show that purification, and its generalization which we call thresholding, lead to significantly stronger play than the standard approach in a wide variety of experimental domains. First, we show that purification improves performance in random 4x4 matrix games using random 3x3 abstractions. We observe that whether or not purification helps in this setting depends crucially on the support of the equilibrium in the full game, and we precisely specify the supports for which purification helps. Next we consider a simplifed version of poker called Leduc Hold'em; again we show that purification leads to a significant performance improvement over the standard approach, and furthermore that whenever thresholding improves a strategy, the biggest improvement is often achieved using full purification. Finally, we consider actual strategies that used our algorithms in the 2010 AAAI Computer Poker Competition. One of our programs, which uses purification, won the two-player no-limit Texas Hold'em bankroll division. Furthermore, experiments in two-player limit Texas Hold'em show that these performance gains do not necessarily come at the expense of worst-case exploitability and that our algorithms can actually produce strategies with lower exploitabilities than the standard approach.