Utility algorithms

Production Utility

Wed, 10/07/2009 - 12:30 — terry

Setting the utility of a production

By default, all productions start with a utility of 0. If you want to set this utility manually, rather than using a learning rule, you can do it like this:

def attendProbe(goal='state:start',vision='busy:False',location='?x ?y',utility=0.3):

Utility Learning Rules

The following learning rules exist for Python ACT-R

• • PMPGC: the old PG-C learning rule

pm=PMPGC(goal=20)

• • PMPGCSuccessWeighted: the success-weighted PG-C rule

pm=PMPGCSuccessWeighted(goal=20)

• • PMPGCMixedWeighted: the mixed-weighted PG-C rule

pm=PMPGCMixedWeighted(goal=20)

• • PMQLearn: standard Q-learning

pm=PMQLearn(alpha=0.2,gamma=0.9,initial=0)

• • PMTD: TD-Learning (Fu & Anderson, 2004)

pm=PMQLearn(alpha=0.1,discount=1,cost=0.05)

• • PMNew: the new ACT-R 6 standard learning rule

pm=PMNew(alpha=0.2)