Heuristic Speed-Ups for Learning in Complex Stochastic Environments

Abstract

We describe a novel methodology by which a software agent can learn to predict future events in complex stochastic environments together with an important heuristic-based acceleration technique for computing the prediction. This speed-up enables us to use much more context in our predictions than was previously possible [Darken, 2005]. We present results gathered from a first prototype of our approach.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 2005
Accession Number
ADA575048

Entities

People

  • Christian J. Darken

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Autonomy
  • Weapons Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Applied Computer Science
  • Artificial Intelligence
  • Artificial Intelligence Software
  • Computational Science
  • Computer Science
  • Environment
  • Information Processing
  • Information Systems
  • Learning
  • Machine Learning
  • Markov Models
  • Models
  • Predictive Modeling
  • Probability
  • Probability Distributions
  • Software Agents

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.
  • Parallel and Distributed Computing.