Home > marl > agent > learnfuns > distrq.m

distrq

PURPOSE ^

Implements the distributed Q-learning algorithm

SYNOPSIS ^

function a = distrq(a, state, action, reward, params)

DESCRIPTION ^

Implements the distributed Q-learning algorithm
  A = TEAMQ(A, STATE, ACTION, REWARD, PARAMS)
  Implements distributed Q-learning. 

  Required values on the agent learning parameters:
   gamma           - the discount factor
   epsilon         - the exploration probability
 
  Required values on the extra parameters:
   newtrial        - only in episodic environments; whether a new trial is
                   beginning

 Supports discrete states and actions, with 1 action variable per agent.

 Should be coupled with determpiact.


  References:
  [1] Lauer, M. & Riedmiller, M. An Algorithm for Distributed
  Reinforcement Learning in Cooperative Multi-Agent Systems Proceedings
  17th International Conference on Machine Learning (ICML-00), 2000,
  535-542

  See also agent_learn, distrq_init

CROSS-REFERENCE INFORMATION ^

This function calls: This function is called by:
Generated on Wed 04-Aug-2010 16:55:08 by m2html © 2005