Opponent Modeling and Spatial Similarity to Retrieve and Reuse Superior Plays

Abstract

Plays are sequences of actions to be undertaken by a collection of agents, or teammates. The success of a play depends on a number of factors including, perhaps most importantly, the opponent's play. In this paper, we present an approach for online opponent modeling and illustrate how it can be used to improve offensive performance in the Rush 2008 football simulator. In football, team behaviors have an observable spatio-temporal structure, defined by the relative physical positions of team members over time. We demonstrate that this structure can be exploited to recognize football plays at a very early stage. Using the recognized defensive play, knowledge about expected outcomes, and spatial similarity between offensive plays, we retrieve an offensive play from the case base. This play is then (partially) reused to improve an in-progress offensive play. We call this process a play switch. Empirical results indicate that spatial similarity is central to play retrieval, and that substituting only a subset of the current play yields greater improvement over a full play substitution.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2009
Accession Number
ADA593111

Entities

People

  • David W. Aha
  • Gita Sukthankar
  • Kennard Laviers
  • Matthew Klenk
  • Matthew Molineaux

Organizations

  • Knexus Research (United States)

Tags

Communities of Interest

  • Autonomy
  • Energy and Power Technologies
  • Human Systems

DTIC Thesaurus Topics

  • Algorithms
  • Computers
  • Information Operations
  • Instructions
  • Kernel Functions
  • Machine Learning
  • Observation
  • Reasoning
  • Recognition
  • Reinforcement Learning
  • Sequences
  • Simulations
  • Simulators
  • Supervised Machine Learning
  • Switches
  • Switching
  • Training

Readers

  • Database Systems and Applications
  • Game Theory.
  • Neural Network Machine Learning.