Re: What did that thread indicate?

"Glen M. Sizemore" <gmsizemore2@xxxxxxxxx> wrote:

>
>"Traveler" <traveler@xxxxxxxxxx> wrote in message
>news:t9dli1p50av1kdt2h6orqgh6jd7emjphar@xxxxxxxxxx
>> On 15 Sep 2005 15:53:17 -0700, humiguel@xxxxxxx wrote:
>>
>>>A reinforcement-learning machine can, at least in principle,
>>>generate secondary behaviour by combining primary behaviours,
>>>testing that in the environment and learning from the feedback.
>>
>> I agree but I would not use the word 'generate'. Reinforcement
>> learning is a selection mechanism. It does not generate anything. It
>> assumes the a priori existence of behaviors. It is used for combining
>> low-level behaviors into high-level apetitive and aversive behaviors.
>> A different mechanism and principle (motor coordination) is
>> responsible for actually generating/creating the behaviors.
>
>
>One thing is certain, using "shaping" (differential reinforcement of
>successive approximations) animals will acquire behavior that would never
>have otherwise been observed.

How can you ever be sure of that? There are many behaviours that
exist in latent form and only manifest themselves when the right
conditions are met. I'm sure that even if you were never seen
killing someone, you'll do that if the right conditions arise. And
if you enlist in the army, you'll be trained to do precisely that.

Anyway, what kind of behaviours are you talking about? For instance
can you train a pigeon to stand belly-up, with its eyes closed as if
it were dead?

In any case the situation were you have a human manipulating the
environment to achieve some goal is very different from the one
where a machine must learn from the feedback provided by the
environment to its own actions.


Antonio Esteves

--
Corby - A new approach to Artificial Intelligence
http://futalgo.planetaclix.pt/corby/index.htm

.

Relevant Pages

  • Re: Is sentience an emergent brain behavior?
    ... The behaviours we produce are initially function of our ... the environment. ... sleepiness to see if the child was ready for bed. ... painterly technique - his parents were sooo worried ...
    (comp.ai.philosophy)
  • Re: Cool visual illusion
    ... Moreover, in humans and other animals, these behaviours _as observed_ are all learned. ... That is, they are capable of some visual behaviours, which are then shaped by the environment in which they exist. ... Assume that you have a visual device that can recognise certain cues, such as a flashing lights, and can discriminate between colours and flash rates. ... If you want to model selective inattenion, you would thus have to model interfering inputs to the visual device. ...
    (comp.ai.philosophy)
  • Re: death of the mind.
    ... > its environment constitute a system. ... > result in observable behaviours, we cannot infer them, since different ... > behaviours may be caused by the same or similar state changes. ...
    (sci.cognitive)
  • Re: Is sentience an emergent brain behavior?
    ... No one ever said that operant conditioning explains 'all' of what you are, you can only condition what's already there. ... Like any animal, we produce behaviours, that is, we interact with our environments. ... But the way they are shaped and develop depends crucially on the environment. ... Because you grew up in an especially liberal period in Western culture, you were trained to question "proper training." ...
    (comp.ai.philosophy)
  • Re: What is conditioning?
    ... This is learning, very rapid, but real learning all the same. ... Learning means the shaping and combining of existing behaviours. ... together as it interacts with its environment. ... Skinner exaggerated the emphasis on conditioning kids to become just about anything you wanted to counteract the idea that people were genetically destined to be one thing or another. ...
    (comp.ai.philosophy)