We describe an approach towards integrating the semantics of semi-structured documents with task-support for (weakly-structured) business processes and proactive inferencing capabilities of a desk support agent. The mechanism of our Proactive Inferencing Agent is motivated by the requirements posed in (weakly-structured) business processes performed by a typical knowledge worker and by experiences we have made from a rst trial with a Reactive Agent Support scheme. Our reactive scheme is an innovative approach for smart task support that links knowledge from an organizational memory to business tasks. The scheme is extended to include proactive inferencing capabilities in order to improve user-friendliness and to facilitate modeling of actual agent support. In particular, the improved scheme copes with varying precision of knowledge found in the organizational memory and it reasons proactively about what might be interesting to the knowledge worker and what might be due in h...

We describe a method for time-critical decision making involving sequential tasks and stochastic processes. The method employs several iterative refinement routines for solving different aspects of the decision making problem. This paper concentrates on the meta-level control problem of deliberation scheduling, allocating computational resources to these routines. We provide different models corresponding to optimization problems that capture the different circumstances and computational strategies for decision making under time constraints. We consider precursor models in which all decision making is performed prior to execution and recurrent models in which decision making is performed in parallel with execution, accounting for the states observed during execution and anticipating future states. We describe algorithms for precursor and recurrent models and provide the results of our empirical investigations to date. 1 Introduction We are interested in solving sequential decision ma...

We describe an approach towards integrating the semantics of semi-structured documents with tasksupport for (weakly-structured) business processes and proactive inferencing capabilities of a desk support agent. The proactive assistance of the intelligent agent is motivated by the requirements posed in (weakly-structured) business processes performed by a typical knowledge worker. First, we introduce a reactive agent that provides knowledge out of an organizational memory for the business task at hand. The building of the reactive agent requires rather rigid query structures that do not t nicely with varying precision of knowledge found in the organizational memory. Thus, we propose an enhanced agent that reasons proactively about what might be interesting to you and what might be due in your next step. Introduction Intelligent information agents are often compared against their human counterparts, such as secretaries or other colleagues. The comparisons usually nd th...

Automation of planning techniques can potentially save a great deal of design and programming time, and can help robots design plans when human help is not available. Currently, the computational cost of machine planning algorithms prevents wide-spread use of these systems, and these costs are magnified in multiagent systems. The most expensive and time-consuming aspect of multiagent systems, communication, must be reduced if multiagent planning is to be practical. We propose a method by which agents may reduce both planning and communication costs by planning with stringent social laws, relaxing the laws as needed to find a solution. We provide and implement a practical model for representing and relaxing social laws, and show a method for learning laws in minimal time; we also present a method for generating and relaxing exclusive resource allocations in specific planning situations, and show that the method performs significantly better than random or no allocation in the average ca...

We present two new algorithms, LCM and IBFS, that make locally optimal incremental planning decisions for the task of finding a path to a goal state in a problem space that contains cycles. Earlier work (RTA* [10]) only solves this locally optimal decision problem when the problem space is tree structured. We precisely characterize the time and space complexity of both new algorithms, and show that they are asymptotically optimal. In addition, we present empirical evidence for a variety of maze problem spaces which shows that LCM and IBFS consistently produce shorter solutions than RTA*, and that the average computation cost of IBFS and LCM grows only slowly with the size of the problem. Finally, we show that as the lookahead depth increases, both algorithms require less computation per move decision than RTA*. An earlier version of this paper appeared in [15]. This research was supported by an NSF Presidential Young Investigator Award, No. IRI-8552925, and a grant from Rockwell Inte...

Partially Observable Markov Decision Process (POMDP) is a general sequential decision-making model where the effects of actions are...

Envelopes are a form of decision rule for monitoring plan execution. We describe one type, the DP envelope, that draws its decisions from a look-up table computed off-line by dynamic programming. Based on an abstract model of agent progress, DP envelopes let a developer approach execution monitoring as a problem independent of issues in agent design. We discuss the application of DP envelopes to a small transportation planning simulation, and discuss the issues that arise in an empirical analysis of the results.

Anytime algorithms are useful when the time available for computation is limited, that is, when there is a tradeoff between the time cost of further computation and the cost of using a solution that is only partially complete. Although machine planning presents this sort of problem, there has not yet been a treatment of anytime planning for the general case, that is, a treatment not tied to specific domains. In this paper, we present a model for general-purpose anytime planning which allows the user to trade off the optimality of plans generated deliberatively with the speed of reactive plan generation. The anytime planner allows an interruption of hierarchical deliberative planning at the completion of any criticality level, and completes the plan at execution time using reactive planning. We illustrate the usefulness of this approach on a manufacturing domain. Introduction Automation of planning techniques can potentially save a great deal of design and programmin...

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1 1.1 Goals and Scope . . . . . . . . . . . . . . . . . . . . . . . . . 1-2 1.2 Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3 II. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1 2.1 Anytime algorithms . . . . . . . . . . . . . . . . . . . . . . . 2-1 2.2 Algorithm Combinations . . . . . . . . . . . . . . . . . . . . 2-3 2.3 Control Theory . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4 2.4 Intelligent Control . . . . . . . . . . . . . . . . . . . . . . . . 2-5 2.5 Bayesian Networks . . . . . . . . . . . . . . . . . . . . . . . . 2-8 2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10 III. Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1 3.1 Phase 1: Architecture Development . . . . . . . . . . . . . . 3-1 ...

this paper we introduce a novel approach to compute nearest neighbour based on a local metric which we call AASM (asymmetric anisotropic similarity metric). In this approach we make two basic assumptions. The first one (anisotropic) states that the metric is defined locally: the space around a trial case is measured using the metric attached to that case. The second one (asymmetric) states that the distance between two points in a continuous feature space F i is not symmetric, i.e.,