Person:

Grosz, Barbara

This paper envisions a multi-agent system that assists patients and their health care providers. This system would support a diverse, evolving team in formulating, monitoring and revising a shared "care plan" that operates on multiple time scales in uncertain environments. It would also enhance communication of health information within this planning framework. The coordination of care for children with complex conditions (CCC), which is a compelling societal need, is presented as a model environment in which to develop and assess such systems. The design of algorithms and techniques needed to realize this vision would yield agents capable of being collaborative partners in health care delivery broadly as well as in other environments with similar properties such as rescue and rebuilding after natural disasters. This paper describes the key characteristics of collaborative health care plan support, defines a set of essential capabilities for autonomous "care-augmenting software agents", and discusses three major multi-agents systems research challenges that building such agents raises: evolving long-term plan management, enhancing team interactions, and leveraging human computation for care plan customization.

This paper studies information exchange in collaborative group activities involving mixed networks of people and computer agents. It introduces the concept of "nearly decomposable" decision-making problems to address the complexity of information exchange decisions in such multi-agent settings. This class of decision-making problems arise in settings which have an action structure that requires agents to reason about only a subset of their partners' actions – but otherwise allows them to act independently. The paper presents a formal model of nearly decomposable decision-making problems, NED-MDPs, and defines an approximation algorithm, NED-DECOP that computes efficient information exchange strategies. The paper shows that NED-DECOP is more efficient than prior collaborative planning algorithms for this class of problem. It presents an empirical study of the information exchange decisions made by the algorithm that investigates the extent to which people accept interruption requests from a computer agent. The context for the study is a game in which the agent can ask people for information that may benefit its individual performance and thus the groupʼs collaboration. This study revealed the key factors affecting peopleʼs perception of the benefit of interruptions in this setting. The paper also describes the use of machine learning to predict the situations in which people deviate from the strategies generated by the algorithm, using a combination of domain features and features informed by the algorithm. The methodology followed in this work could form the basis for designing agents that effectively exchange information in collaborations with people.

This paper addresses the problem of computing the value of information in settings in which the people using an autonomous-agent system have access to information not directly available to the system itself. To know whether to interrupt a user for this information, the agent needs to determine its value. The fact that the agent typically does not know the exact information the user has and so must evaluate several alternative possibilities significantly increases the complexity of the value-of-information calculation. The paper addresses this problem as it arises in multi-agent task planning and scheduling with architectures in which information about the task schedule resides in a separate “scheduler” module. For such systems, calculating the value to overall agent performance of potential new information requires that the system component that interacts with the user query the scheduler. The cost of this querying and inter-module communication itself substantially affects system performance and must be taken into account. The paper provides a decision-theoretic algorithm for determining the value of information the system might acquire, query-reduction methods that decrease the number of queries the algorithm makes to the scheduler, and methods for ordering the queries to enable faster decision-making. These methods were evaluated in the context of a collaborative interface for an automated scheduling agent. Experimental results demonstrate the significant decrease achieved by using the query-reduction methods in the number of queries needed for reasoning about the value of information. They also show the ordering methods substantially increase the rate of value accumulation, enabling faster determination of whether to interrupt the user.

Teamwork and care coordination are of increasing importance to health care delivery and patient safety and health. This paper describes our initial work on developing agents that are able to make intelligent information sharing decisions to support a diverse, evolving team of care providers in constructing and maintaining a shared plan that operates in uncertain environments and over a long time horizon.

This paper describes a challenging plan recognition problem that arises in environments in which agents engage widely in exploratory behavior, and presents new algorithms for effective plan recognition in such settings. In exploratory domains, agentsʼ actions map onto logs of behavior that include switching between activities, extraneous actions, and mistakes. Flexible pedagogical software, such as the application considered in this paper for statistics education, is a paradigmatic example of such domains, but many other settings exhibit similar characteristics. The paper establishes the task of plan recognition in exploratory domains to be NP-hard and compares several approaches for recognizing plans in these domains, including new heuristic methods that vary the extent to which they employ backtracking, as well as a reduction to constraint-satisfaction problems. The algorithms were empirically evaluated on peopleʼs interaction with flexible, open-ended statistics education software used in schools. Data was collected from adults using the software in a lab setting as well as middle school students using the software in the classroom. The constraint satisfaction approaches were complete, but were an order of magnitude slower than the heuristic approaches. In addition, the heuristic approaches were able to perform within 4% of the constraint satisfaction approaches on student data from the classroom, which reflects the intended user population of the software. These results demonstrate that the heuristic approaches offer a good balance between performance and computation time when recognizing peopleʼs activities in the pedagogical domain of interest.

When interacting with computer agents, people make inferences about various characteristics of these agents, such as their reliability and trustworthiness. These perceptions are significant, as they influence people’s behavior towards the agents, and may foster or inhibit repeated interactions between them. In this paper we investigate whether computer agents can use the expression of emotion to influence human perceptions of trustworthiness. In particular, we study human-computer interactions within the context of a negotiation game, in which players make alternating offers to decide on how to divide a set of resources. A series of negotiation games between a human and several agents is then followed by a “trust game.” In this game people have to choose one among several agents to interact with, as well as how much of their resources they will trust to it. Our results indicate that, among those agents that displayed emotion, those whose expression was in accord with their actions (strategy) during the negotiation game were generally preferred as partners in the trust game over those whose emotion expressions and actions did not mesh. Moreover, we observed that when emotion does not carry useful new information, it fails to strongly influence human decision-making behavior in a negotiation setting.

Children with complex health conditions require care from a large, diverse team of caregivers that includes multiple types of medical professionals, parents and community support organizations. Coordination of their outpatient care, essential for good outcomes, presents major challenges. Extensive healthcare research has shown that the use of integrated, team-based care plans improves care coordination, but such plans are rarely deployed in practice. This paper reports on a study of care teams treating children with complex conditions at a major university tertiary care center. This study investigated barriers to plan implementation and resultant care coordination problems. It revealed the complex nature of teamwork in complex care, which poses challenges to team coordination that extend beyond those identified in prior work and handled by existing coordination systems. The paper builds on a computational teamwork theory to identify opportunities for technology to support increased plan-based complex-care coordination and to propose design approaches for systems that enable and enhance such coordination.

This paper defines the "Single Agent in a Team Decision" (SATD) problem. SATD differs from prior multi-agent communication problems in the assumptions it makes about teammates' knowledge of each other's plans and possible observations. The paper proposes a novel integrated logical-decision-theoretic approach to solving SATD problems, called MDP-PRT. Evaluation of MDP-PRT shows that it outperforms a previously proposed communication mechanism that did not consider the timing of communication and compares favorably with a coordinated Dec-POMDP solution that uses knowledge about all possible observations.

This paper considers the design of agent strategies for deciding whether to help other members of a group with whom an agent is engaged in a collaborative activity. Three characteristics of collaborative planning must be addressed by these decision-making strategies: agents may have only partial information about their partners' plans for sub-tasks of the collaborative activity; the effectiveness of helping may not be known a priori; and, helping actions have some associated cost. The paper proposes a novel probabilistic representation of other agents' beliefs about the recipes selected for their own or for the group activity, given partial information. This representation is compact, and thus makes reasoning about helpful behavior tractable. The paper presents a decision-theoretic mechanism that uses this representation to make decisions about two kinds of helpful actions: communicating information relevant to a partner's plans for some sub-action, and adding domain actions that are helpful to other agent(s) into the collaborative plan. This mechanism includes a set of rules for reasoning about the utility of helpful actions and the cost incurred by doing them. It was tested using a multi-agent test-bed with configurations that varied agents' uncertainty about the world, their uncertainty about each others' capabilities or resources, and the cost of helpful behavior. In all cases, agents using the decision-theoretic mechanism to decide whether to help outperformed agents using purely axiomatic rules.

This paper proposes the use of restructuring information about choices to improve the performance of computer agents on recurring sequentially dependent decisions. The intended situations of use for the restructuring methods it defines are website platforms such as electronic marketplaces in which agents typically engage in sequentially dependent decisions. With the proposed methods, such platforms can improve agents’ experience, thus attracting more customers to their sites. In sequentially-dependent-decisions settings, decisions made at one time may affect decisions made later; hence, the best choice at any point depends not only on the options at that point, but also on future conditions and the decisions made in them. This “problem restructuring” approach was tested on sequential economic search, which is a common type of recurring sequentially dependent decision-making problem that arises in a broad range of areas. The paper introduces four heuristics for restructuring the choices that are available to decision makers in economic search applications. Three of these heuristics are based on characteristics of the choices, not of the decision maker. The fourth heuristic requires information about a decision-makers prior decision-making, which it uses to classify the decision-maker. The classification type is used to choose the best of the three other heuristics. The heuristics were extensively tested on a large number of agents designed by different people with skills similar to those of a typical agent developer. The results demonstrate that the problem-restructuring approach is a promising one for improving the performance of agents on sequentially dependent decisions. Although there was a minor degradation in performance for a small portion of the agents, the overall and average individual performance improved substantially. Complementary experimentation with people demonstrated that the methods carry over, to some extent, also to human decision makers. Interestingly, the heuristic that adapts based on a decision-maker’s history achieved the best results for computer agents, but not for people.