One of the challenges of multiagent decision making is that the behavior needed to maximize utility can depend on what other agents choose to do: sometimes there is no "right" answer in the absence of knowledge of how opponents will act. The Nash equilibrium is a sensible choice of behavior because it represents a mutual best response. But, even when there is a unique equilibrium, other players are under no obligation to take part in it. This observation has been forcefully illustrated in the behavioral economics community where repeated experiments have shown individuals playing Nash equilibria and performing badly as a result. In this paper, we show how to apply a tool from behavioral economics called the Cognitive Hierarchy (CH) to the design of agents in general sum games. We attack the recently introduced ``Lemonade Game'' and show how the results of an open competition are well explained by CH. We believe this game, and perhaps many other similar games, boils down to predicting how deeply other agents in the game will be reasoning. An agent that does not reason enough risks being exploited by its opponents, while an agent that reasons too much may not be able to interact productively with its opponents. We demonstrate these ideas by presenting empirical results using agents from the competition and idealizations arising from a CH analysis.

In this paper we study the sequential strategic interactive setting of two-person, two-stage, seller-offers bargaining under uncertainty. We model the epistemology of the problem in a finite interactive decision-theoretic framework and solve it for three types of agents of successively increasing (epistemological) sophistication (or, capacity to represent and reason with higher orders of beliefs). In particular, we remove common knowledge assumptions about the agents' epistemology which, if made, would be sufficient to imply the existence of a, possibly unique, game-theoretic equilibrium solution. In this context, we present a characterization of a monotonic relationship between an agent's optimal behavior and its beliefs under a particular moment-based ordering. Further, based on this characterization, we present the \emph{spread-accumulate} sampling technique -- a method of sampling an agent's higher order belief by generating ``evenly dispersed" beliefs for which we (pre)compute offline solutions. Then, we present a method for approximating higher order prior belief update to arbitrary precision by identifying a (previously solved) belief ``closest" to the true belief. In addition, these methods directly suggest a mechanism for achieving a balance between efficiency and the quality of the approximation -- either by generating a large number of offline solutions or by allowing the agent to search online for a ``closer" belief in the vicinity of best current solution.

It is crucial for social systems to adapt to the dynamics of open environments. This adaptation process becomes especially challenging in the context of multiagent systems. In this paper, we argue that multiagent meta-level control is an effective way to determine when this adaptation process should be done and how much effort should be invested in adaptation as opposed to continuing with the current action plan. We develop a reinforcement learning based mechanism for multiagent meta-level control that facilitates the metalevel control component of each agent to learn policies in a decentralized fashion that (a) it can efficiently support agent interactions with other agents and (b) reorganize the underlying network when needed. We evaluate this mechanism in the context of a multiagent tornado tracking application called NetRads. Empirical results show that adaptive multiagent meta-level control significantly improves the performance of the tornado tracking network for a variety of weather scenarios.

Informational conflicts in operational policies cause agents to run into situations where responding based on the rules in one policy violates the same or another policy. Static checking of these conflicts is infeasible and impractical in a dynamic environment. This paper discusses a practical approach to handling policy conflicts in real-time domains within the context of a hierarchical military command and control simulated system that consists of a central command, squad leaders and squad members. All the entities in the domain function according to preset communication and action protocols in order to perform successful missions. Each entity in the domain is equipped with an instance of a metacognitive component to provide on-board/on-time analysis of actions and recommendations during the operation of the system. The metacognitive component is the Metacognitive Loop (MCL) which is a general purpose anomaly processor designed to function as a cross-domain plugin system. It continuously monitors expectations and notices when they are violated, assesses the cause of the violation and guides the host system to an appropriate response. MCL makes use of three ontologies—indications, failures and responses—to perform the notice, assess and guide phases when a conflict occurs. Conflicts in the set of rules (within a policy or between policies) manifest as expectation violations in the real world. These expectation violations trigger nodes in the indication ontology which, in turn, activate associated nodes in the failure ontology. The responding failure nodes then activate the appropriate nodes in the response ontology. Depending on which response node gets activated, the actual response may vary from ignoring the conflict to prioritizing, modifying or deleting one or more conflicting rules.

One of the critical factors for a successful cooperative relationship in a supply chain partnership is trust. Many real-world applications, such as Supply Chain Management (SCM), can be modeled using multi-agent systems. One shortcoming of current SCM models is that their trust models are ad hoc and do not have a strong theoretical basis. As a result, they are unable to model subtleties in agent behavior that can be used to build a more accurate trust model. We propose a trust model for SCM that is grounded in probabilistic game theory. In this model, trust can be gained through direct interactions and/or by asking for information from other trustworthy agents. We will use this model to simulate and study supply chain market behavior.

The most common solution concept for a strategic interaction situation is the Nash equilibrium, in which no agent can do better by deviating unilaterally. However, the Nash equilibrium underlays on the assumption of common information that is hardly verified in many practical situations. When information is not common, rational agents are assumed to learn from their observations to derive beliefs over their opponents' play and payoffs. In these situations, there are steady states composed of beliefs and strategies in which the strategies do not constitute a Nash equilibrium. These stable states are called in the game theory literature self-confirming equilibria. They are such that every agent plays the best response to her beliefs and these are correct on the equilibrium path, while off the equilibrium path they may be incorrect. We present some mathematical programming formulations for computing self-confirming equilibria and its refinements in two-player extensive-form games and we study their properties.

We study several classes of interactive assistants from the points of view of decision theory and computational complexity. We first introduce a special class of POMDPs called hidden-goal MDPs (HGMDPs), which formalize the problem of interactively assisting an agent whose goal is hidden and whose actions are observable. In spite of its restricted nature, we show that optimal action selection in finite horizon HGMDPs is PSPACE-complete even in domains with deterministic dynamics. We then introduce a more restricted model called helper action MDPs (HAMDPs), where the assistant’s action is accepted by the agent when it is helpful, and can be easily ignored by the agent otherwise. We show classes of HAMDPs that are complete for PSPACE and NP along with a polynomial time class. Furthermore, we show that for general HAMDPs a simple myopic policy achieves a regret, compared to an omniscient assistant, that is bounded by the entropy of the initial goal distribution. A variation of this policy is also shown to achieve worst-case regret that is logarithmic in the number of goals for any goal distribution.

Collaborative filtering systems have been developed to manage information overload and improve discourse in online communities. In such systems, users rank content provided by other users on the validity or usefulness within their particular context. The goal is that "good" content will rise to prominence and "bad" content will fade into obscurity. These filtering mechanisms are not well-understood and have known weaknesses. For example, they depend on the presence of a large crowd to rate content, but such a crowd may not be present. Additionally, the community's decisions determine which voices will reach a large audience and which will be silenced, but it is not known if these decisions represent "the wisdom of crowds" or a "censoring mob." Our approach uses statistical machine learning to predict community ratings. By extracting features that replicate the community's verdict, we can better understand collaborative filtering, improve the way the community uses the ratings of their members, and design agents that augment community decision-making. Slashdot is an example of such a community where peers will rate each others' comments based on their relevance to the post. This work extracts a wide variety of features from the Slashdot metadata and posts' linguistic contents to identify features that can predict the community rating. We find that author reputation, use of pronouns, and author sentiment are salient. We achieve 76% accuracy predicting community ratings as good, neutral, or bad.

Wikis allow for collaborators to collect information about entities. In turn, such entity information can be used for AI tasks, such as information extraction. However, these collaborators are almost exclusively human users. Allowing arbitrary software agents to act as collaborators can greatly enrich a wiki since agents can contribute structured data to complement the human-contributed, unstructured-data. For instance, agents can import huge volumes of structured data about entities, enriching the pages, and agents can update wiki pages to reflect real-time information changes (e.g., win-loss records in sports). This paper describes an approach that allows for both arbitrary software agents and human users to collaborate. In particular, we address three key problems: agents updating the correct wiki pages, policies for agent updates, and sharing the schema across collaborators. Using our approach, we describe creating entity-focused wikis which include the ability to create dynamic categories of entities based on their wiki pages. These categories dynamically update their membership based upon real-world changes.

A distributed non-hierarchical metacognitive architec- ture is one in which all meta-level reasoning compo- nents are subject to meta-level monitoring and manage- ment by other components. Such metacognitive distri- bution can support the robustness of distributed IT sys- tems in which humans and artificial agents are partic- ipants. However, robust metacognition also needs to be context-aware and use diversity in its reasoning and analysis methods. Both these requirements mean that an agent evaluates its reasoning within a “bigger picture” and that it can monitor this global picture from multi- ple perspectives. In particular, social context-awareness involves understanding the goals and concerns of users and organisations. In this paper, we first present a conceptual architecture for distributed metacognition with context-awareness and diversity. We then consider the challenges of apply- ing this architecture to autonomic management systems in scenarios where agents must collectively diagnose and respond to errors and intrusions. Such autonomic systems need rich semantic knowledge and diverse data sources in order to provide the necessary context for their metacognitive evaluations and decisions.