REPETITIVE PLAY OF AN UNKNOWN GAME AGAINST NATURE

Abstract

A repetitive play of a game against Nature is considered under the assumption that the player knows nothing about the game except his own set of strategies. After each play, he is told the value of the random loss incurred by him. A strategic rule for the player is defined with the property that the average loss achieves asymptotically the minimum functional of the game in probability and uniformly in all sequences of Nature's strategies. The rate of convergence of expected average losses is shown as well.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 1967
Accession Number
AD0826612

Entities

People

  • Bruno O. Subert

Organizations

  • Stanford University

Tags

Communities of Interest

  • Advanced Electronics

DTIC Thesaurus Topics

  • Air Force
  • Convergence
  • Distribution Functions
  • Electronics
  • Electronics Laboratories
  • Game Theory
  • Hypotheses
  • Inequalities
  • Numbers
  • Probability
  • Random Variables
  • Sequences
  • Sequential Games
  • Statistics
  • Theorems
  • Theses

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Mathematics or Statistics
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.