Jianhui Wu

Solving large TÆMS problems efficiently by selective exploration and decomposition

TÆMS is a hierarchical modeling language capable of representing complex task networks with intra-task uncertainties and inter-task dependencies. The uncertainty and complexity of the application domains represented in TÆMS models often lead to very large state spaces, which push the need to design efficient solution algorithms for TÆMS problems. In this paper, we present a solver that integrates selective state space search techniques with state space decomposition techniques. Our experiments demonstrate that the solver can find an (approximately) optimal solution much faster than prior approaches.

Automated resource-driven mission phasing techniques for constrained agents

A constrained agent is limited in the actions that it can take at any given time, and a challenging problem is to design policies for such agents to do the best they can despite their limitations. One way of improving agent performance is to break larger tasks into phases, where the constrained agent is better able to handle each phase and can reconfigure its limited capabilities differently for each phase. In this paper, we present algorithms for automating the process of finding and using mission phases for constrained agents. We analyze several variations of this problem that correspond to different classes of important constrained-agent problems, and show through analysis and experiments that our techniques can increase an agents rewards for varying levels of constraints on the agent and on the phases.

Resource-driven mission-phasing techniques for constrained agents in stochastic environments

Resource constraints restrict the set of actions that an agent can take, such that the agent might not be able to perform all its desired tasks. Computational time limitations restrict the number of states that an agent can model and reason over, such that the agent might not be able to formulate a policy that can respond to all possible eventualities. This work argues that, in either situation, one effective way of improving the agents performance is to adopt a phasing strategy. Resource-constrained agents can choose to reconfigure resources and switch action sets for handling upcoming events better when moving from phase to phase; time-limited agents can choose to focus computation on high-value phases and to exploit additional computation time during the execution of earlier phases to improve solutions for future phases. This dissertation consists of two parts, corresponding to the aforementioned resource constraints and computational time limitations. The first part of the dissertation focuses on the development of automated resource-driven mission-phasing techniques for agents operating in resource-constrained environments. We designed a suite of algorithms which not only can find solutions to optimize the use of predefined phase-switching points, but can also automatically determine where to establish such points, accounting for the cost of creating them, in complex stochastic environments. By formulating the coupled problems of mission decomposition, resource configuration, and policy formulation into a single compact mathematical formulation, the presented algorithms can effectively exploit problem structure and often considerably reduce computational cost for finding exact solutions. The second part of this dissertation is the design of computation-driven mission-phasing techniques for time-critical systems. We developed a new deliberation scheduling approach, which can simultaneously solve the coupled problems of deciding both when to deliberate given its cost, and which phase decision procedures to execute during deliberation intervals. Meanwhile, we designed a heuristic search method to effectively utilize the allocated time within each phase. As illustrated in analytical and experimental results, the computation-driven mission-phasing techniques, which extend problem decomposition techniques with the across-phase deliberation scheduling and inner-phase heuristic search methods mentioned above, can help an agent judiciously emphasize high-value portions of a large problem, while paying less attention to others, to generate a better policy within its time limit.

Coordination of Highly Contingent Plans

Conventional military planning systems construct plans with very limited flexibility. In the future, military plans will evolve into a much more expressive, contingent form. This paper describes how Honeywells distributed coordinator agents reason about complex domains to construct and execute highly contingent plans. The agents operate in a very dynamic environment in which complex hierarchical tasks can arrive unpredictably and the agents have to build coordinated joint plans on the fly, while execution proceeds. Using carefully limited forms of inter-agent communication, the agents develop agreements on their future coordinated behavior and rely on those agreements to build highly contingent plans (partial policies) that specify what actions they should take in a wide variety of possible futures. As mission execution proceeds and the tasks yield varying outcomes, the agents must rapidly, continually coordinate and adapt their plans. The result is a distributed multi-agent system capable of building and flexibly executing complex, highly-contingent coordinated mission plans

Mathematical programming for deliberation scheduling in time-limited domains

Deliberation scheduling is the process of scheduling decision procedures to maximize overall performance. In this paper, we present novel mathematical programming algorithms for scheduling deliberations, and illustrate them through several increasingly complex classes of deliberation scheduling problems. In comparison to previous work, our algorithms are able to optimally or near-optimally solve deliberation scheduling problems for decision procedures that have more general performance profiles, and are applicable in complex domains with uncertainty. We also illustrate that, thanks to the mathematical programming form illation, our algorithms can be easily extended to model additional system aspects.

The Impact of Communication Costs and Limitations on Price Wars in an Information Economy

Price wars ¿ the iterative undercutting of prices to the marginal cost by competitors ¿ have frequently emerged in models of economic systems populated by computational agents. In this paper, we explore the prevalence and severity of price wars in models of multiagent ecommerce systems that include costs and limitations on interagent communication. The empirical results we describe in this paper indicate that, for a stationary consumer population, limiting the rate of penetration of price information can reduce the severity of price wars, and that charging producer agents for communication can in fact curtail price-undercutting before prices (and profits) bottom out. Furthermore, prices (and profits) do not bottom out for non-stationary consumer populations, where in fact cyclic price wars can arise.