Home > marl > agent > actfuns > staticact.m

staticact

PURPOSE ^

Implements some static action selection algorithms

SYNOPSIS ^

function [a, action] = staticact(a, state, actions, rewards, params)

DESCRIPTION ^

Implements some static action selection algorithms
  [A, ACTION] = STATICACT(A, STATE, ACTIONS, REWARDS, PARAMS)
  Implements a set of static action selection algorithms. They are
  selected via the following learning parameters:
   'static.type'  - either one of the following:
       'fixed'         - a fixed number (the default)
       'noisy'         - a uniformly distributed random number with the 
           mean around a given number
       'rand'          - random choice between discrete actions.
           Naturally, actions must be discrete
       'sequence'      - a predefined sequence of actions (e.g., derived
           from the optimal policy)
   'static.value'   - the fixed value for 'fixed', mean for 'noisy',
           the sequence of actions for 'sequence'.
           Default 0
   'static.stdev'     - noise standard deviation, for 'noisy'. Default
           0.1.


  See also agent_act

CROSS-REFERENCE INFORMATION ^

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