Arrow Research search

Author name cluster

R. Michael Young

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

10 papers
2 author rows

Possible papers

10

ICAPS Conference 2024 Conference Paper

The Story So Far on Narrative Planning

  • Rogelio E. Cardona-Rivera
  • Arnav Jhala
  • Julie Porteous
  • R. Michael Young

Narrative planning is the use of automated planning to construct, communicate, and understand stories, a form of information to which human cognition and enaction is pre-disposed. We review the narrative planning problem in a manner suitable as an introduction to the area, survey different plan-based methodologies and affordances for reasoning about narrative, and discuss open challenges relevant to the broader AI community.

Details

AAAI Conference 2004 Conference Paper

Natural Language Processing and Information Extraction Comparing Cognitive and Computational Models of Narrative Structure

  • David B. Christian
  • R. Michael Young

A growing number of applications seek to incorporate automatically generated narrative structure into interactive virtual environments. In this paper, we evaluate a representation for narrative structure generated by an automatic planning system by 1) mapping the plans that control plot into conceptual graphs used by QUEST, an existing framework for question-answering analysis that includes structures for modeling a reader's narrative comprehension and 2) using methods originally employed by QUEST’s developers to determine if the plan structures can serve as effective models of the understanding that human users form after viewing corresponding stories played out within a virtual world. Results from our analysis are encouraging, though additional work is required to expand the plan language to cover a broader class of narrative structure.

PDF Details

AAAI Conference 1999 Conference Paper

Cooperative Plan Identification: Constructing Concise and Effective Plan Descriptions

  • R. Michael Young
  • North Carolina State University

Intelligent agents are often called uponto formplans that direct their ownor other agents’ activities. For these systems, the ability to describe plans to people in natural waysis an essential aspect of their interface. In this paper, wepresent the CooperativePlanIdentification (CPI) architecture, a computationalmodelthat generatesconcise, effective textual descriptions of plan data structures. Themodelincorporates previous theoretical workon the comprehension of plan descriptions, using a generate-and-test approachto performefficient search throughthe space of candidate descriptions. Wedescribe an empirical evaluation of the CPIarchitecture in whichsubjects following instructions produced by the CPI architecture performed their tasks with fewer execution errors and achieveda higher percentageof their tasks’ goals than did subjects following instructions producedby alternative methods.

PDF Details

AAAI Conference 1996 Conference Paper

Using Plan Reasoning in the Generation of Plan Descriptions

  • R. Michael Young

Previous work on the generation of natural language descriptions of complex activities has indicated that the unwieldy amount of text needed to describe complete plans makes for ineffective and unnatural descriptions. We argue here that concise and effective text descriptions of plans can be generated by exploiting a model of the hearer’ s plan reasoning capabilities. We define a computational model of the hearer’ s interpretation process that views the interpretation of plan descriptions as refinement search through a space of partial plans. This model takes into account the hearer’ s plan preferences and the resource limitations on her reasoning capabilities to determine the completed plans she will construct from a given partial description.

PDF Details

ICAPS Conference 1994 Conference Paper

Decomposition and Causality in Partial-order Planning

  • R. Michael Young
  • Martha E. Pollack
  • Johanna D. Moore

Wedescribe DPOCL, a partinl-order csnsal llnk planner that includes action decomposition. DPOCL builds directly on the SNLPalgorithm (McAllester Rosenbiltt 1991), and hence is clear and simple, ud can readily be integrated with other SNLPextensions. In addition, DPOCL is specifically designed to handle partially specified action decompositions. Plan generation in DPOCL exploits the planner’s ability to fill in the missing pieces of n partially specified subplan in s way that uses the existing context of the larger plan being constructed.

Details