>> More generally, Bayes rule is not an escape from the requirement that we
>> reason non-circularly. If you construct a hypothesis that is
>> self-reinforcing, then you have constructed template which can be used to
>> justify anything.
>
>What do you mean by a "self-reinforcing" hypothesis?
Loose talk for which I apologize.
The hypothesis itself cannot be self-reinforcing unless it is combined
with data and prior assumptions. For example, I might have the prior
assumption that if the tooth fairy existed, she would predispose people
to ask questions about the tooth fairy (more than they would be
predisposed to do otherwise). Now I entertain the hypothesis that the
tooth fairy exists. The fact that I have asked the question is evidence
that I am predisposed to ask the question. This increases my posterior
on the existence of the tooth fairy. Indeed, every time I ask the
question my posterior increases. This is clearly bogus since I can use
this line of reasoning to justify anything. The hypothesis, along with
the other assumptions I have made form a vicious circle.
>But what if the truth isn't among the alternatives you are considering? One of
>the strengths of the Bayesian approach is that it forces you to explicitly
>specify the set of alternatives you are considering. This brings hidden
>assumptions out into the open, and makes it easier to step back and question
>those assumptions... then perhaps redo your analysis taking into account
>additional hypotheses that you didn't consider at first.
Yes. This is a strength. What gets people confused is when the
hypotheses in question pertain to the use or applicability of the rules
of reasoning. This is how circularity can slip in.
--
Ron Parr email: [EMAIL PROTECTED]
--------------------------------------------------------------------------
Home Page: http://robotics.stanford.edu/~parr
