Towards a Plan-based Learning Environment

Proceedings of the 35th Annual ACM Symposium on Applied Computing, 2020

Scroll to the title and select a Blue link to open a paper. After viewing the paper, use the bookmarks to the left to return to the beginning of the Table of Contents.

Lecture Notes in Computer Science, 2004

This paper presents the application of a novel method in the multiagent teamwork field called Scenario-based Teamworking (SBT). In SBT method a team of cooperative intelligent agents could be able to execute complex plans in nondeterministic, adversary, and dynamic environments which communication cost is high. The base idea of this method is to define Scenario for different situations. With a graph of scenarios, a team of agents can execute, learn, adapt, and create team plans automatically. This method has implemented in a soccer team of intelligent agents (players and coach) and evaluated in the standard RoboCup simulator environment [1] and results show a significant improvement.

Reliability Engineering & System Safety, 1988

Modelling (ASM) provides a possible frame of reference .for monitoring operator activities, but it cannot be used directly: (1) operators may deviate .from the scenario described in A SM, (2) the actual situation may develop d(fferently from the scenario, (3) operators are normally involved in several activities at the same time, and (4) modelling of operator activities must focus on the level of individual actions, while the ASM only addresses the global view. The reference provided by the ASM scenario must therefore be supplemented by a more direct modelling of what the operator does. This requires a recognition of the operator's current plans, i.e. his goals and the strategies he employs to reach them. The paper describes an ongoing programme to develop an expert system that does this, within the ESPRIT project Graphical Dialogue Environment.

Proceedings of the Fifteenth …, 1999

We present a new abductive, probabilistic theory of plan recognition. This model differs from previous theories in being centered around a model of plan execution: most previous methods have been based on plans as formal objects or on rules describing the recognition process. We show that our new model accounts for phenomena omitted from most previous plan recognition theories: notably the cumulative effect of a sequence of observations of partially-ordered, interleaved plans and the effect of context on plan adoption. The model also supports inferences about the evolution of plan execution in situations where another agent intervenes in plan execution. This facility provides support for using plan recognition to build systems that will intelligently assist a user.

Advances in Artificial Intelligence, 1996

There has been great interest in the formalization and development of systems that treat plan recognition process in Artificial Intelligence. In this paper, we develop a theoretical framework, and implement a plan recognition system. Therefore, our research work can be divided in two parts: a hybrid theory, and a feasible implementation. In the theoretical part, we develop a Hybrid Formal Theory—HFT, based in the theory of Kautz and other researchers. In the feasible part, we create a tool called Plan Recognition System—PRS.

1999

Abstract. The output of advice-giving systems can be regarded as plans to be executed by the user. Such plans are fairly useless if the user is not capable of executing some of the involved plan steps, or if he does not know them. We propose a two-phase process of user-tailored plan generation and plan presentation to produce advice that enables a user to reach his goals. This paper reports the first phase, the generation of a plan under the constraints of the user's capabilities.

Lecture Notes in Computer Science, 1995

This research examines the metaphor of goal-driven planning as a tool for performing the integration of multiple learning algorithms. In case-based reasoning systems, several learning techniques may apply to a given situation. In a failure-driven learning environment, the problems of strategy construction are to choose and order the best set of learning algorithms or strategies that recover from a processing failure and to use those strategies to modify the system's background knowledge so that the failure will not be repeated in similar future situations. A solution to this problem is to treat learning-strategy construction as a planning problem with its own set of goals. Learning goals, as opposed to ordinary goals, specify desired states in the background knowledge of the learner, rather than desired states in the external environment of the planner. But as with traditional goalbased planners, management and pursuit of these learning goals becomes a central issue in learning. Example interactions of learning-goals are presented from a multistrategy learning system called Meta-AQUA that combines a case-based approach to learning with nonlinear planning to achieve goals in a knowledge space. To appear in M. Keane, J.-P. Haton, & M. Manago (Eds.), Topics in case-based reasoning (Lecture notes in artificial intelligence). Berlin: Springer-Verlag. 1. Explanation-Pattern Theory has roots in the work of Schank [32] and Ram . See Ram and Cox [26] and Cox [4] for specific details on the algorithm used to retrieve and utilize Meta-XPs. The algorithm is adapted from standard XP algorithms. See also [13] and [33] for additional uses of XP theory.

2002

This paper presents our evolving view of organizational p anning. We discuss lessons we are learning by studying examples of real-world organizational planning. Our approach to this topic is interdisciplinary and founded on a view of planning as a form of resource-constrained d cision making under uncertainty. We xamine how considerations of planning as a multi-agent process dictates extensions to our basic decision-theoretic perspective by examining some of the challenges of coordinating planning activities and attaining coherence in the contents of plans constructed by multiple, independent planners. We then consider the multiple uses that organizational plans seem to serve in human organizations. We conclude by summarizing some implications of our studies for the development of organizational planning technology.

1997

This paper discusses how a plan execution system could be used to assist in a knowledge management task, such as searching a set of databases for a relevant document. PARETO iS a plan execution system based on the as-needed expansion of a hierarchy of sketchy plans. It has been adapted for use in several different domains. There are sev- eral problems that would arise in adapting PARETO to assist in a knowledge management task such as searching for a relevant document in a proprietary intranet. However, there are also advantages that would be provided by the use of such a system.

ppml.url.tw

The objective of this article is to contribute to creating a better understanding of the Last Planner System (LPS)-which is associated with Lean Construction-in the light of the learning processes at the basis of knowledge development, and of change and innovation. Founded on a theoretical discussion, three research questions are asked, namely: In what ways can the LPS be expected to alter the learning arenas compared to conventional project management in construction; according to learning theory, what are the main challenges associated with implementing the LPS; and, finally, what kind of learning can be linked to an implemented LPS that functions as intended? The implementation of the LPS is shown to require substantial changes to the technical-organisational learning arena. In order for the implementation to be successful, the work identity has to alter on the individual level so that an overlap occurs with the new work practices prescribed by the LPS. The LPS has an inbuilt experiential learning cycle, and provides a good starting point for single-loop learning, as well as for simple forms of double-loop learning ("routinized learning capability"). However, it is argued that the LPS understood as experiential learning has clear limitations with regard to "evolutionary learning capability". This is amplified by the context project organisation provides. In terms of theoretical implications, this article promotes an understanding of the planning process informed by the theory describing it as an experiential learning cycle. The conceptualisation which separates the LPS from conventional production control theory is critiqued. Finally, it is argued that an understanding of the LPS grounded in learning theory will improve the possibilities for successful implementation and maximise the learning effects.