[x3d-public] Off-subject: suppression learning
cbullard at hiwaay.net
cbullard at hiwaay.net
Fri Mar 12 18:14:06 PST 2021
On 2021-03-12 18:26, John Carlson wrote:
> So what’s the opposite of reinforcement learning? Insuring that
> the ai player of a video game receives a low score? The lower the
> score, the better for the game ai?
> _______________________________________________
> x3d-public mailing list
> x3d-public at web3d.org
> http://web3d.org/mailman/listinfo/x3d-public_web3d.org
In classical Skinnerian paradigms, it is negative reinforcement. It is
used to
extinguish a behavior if the behavior should be extinguished quickly.
It is
usually advised one not use this technique. The pigeon will peck you
and
the rat will bite you if used too often. It makes the subject mean. a
better technique is to use positive reinforcement to train the subject
away
from the adverse behavior to the desired behavior.
len
More information about the x3d-public
mailing list