Learning Continuous Action Models in a Real-Time Strategy Environment

Abstract

Although several researchers have integrated methods for reinforcement learning (RL) with case-based reasoning (CBR) to model continuous action spaces, existing integrations typically employ discrete approximations of these models. This limits the set of actions that can be modeled, and may lead to non-optimal solutions. We introduce the Continuous Action and State Space Learner (CASSL), an integrated RL/CBR algorithm that uses continuous models directly. Our empirical study shows that CASSL significantly outperforms two baseline approaches for selecting actions on a task from a real-time strategy gaming environment.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2008
Accession Number
ADA593081

Entities

People

  • David W. Aha
  • Matthew Molineaux
  • Philip Moore

Organizations

  • Knexus Research (United States)

Tags

Communities of Interest

  • Autonomy
  • Counter WMD
  • Human Systems
  • Weapons Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence
  • Artificial Intelligence Computing
  • Computational Complexity
  • Computational Science
  • Computers
  • Environment
  • Information Science
  • Intelligent Agents
  • Kernel Functions
  • Learning
  • Machine Learning
  • Neural Networks
  • Reasoning
  • Reinforcement Learning
  • Simulations
  • Training

Fields of Study

  • Computer science

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • Space