On Jan 28, 2008 6:40 AM, Ben Goertzel <[EMAIL PROTECTED]> wrote: > Nesov wrote: > > Exactly. It needs stressing that probability is a tool for > > decision-making and it has no semantics when no decision enters the > > picture. > ... > > What's it good for if it can't be used (= advance knowledge)? For > > other purposes we'd be better off with specially designed random > > number generators. So it's more like tautology that anything useful > > influences decisions. > > > In another context, I might not be picky about the use of the word > "decision" here ... but this thread started with a discussion of radical > models of the universe involving multi-multiverses and Yverses > and so on. > > In this context, casual usage of folk-psychology notions like "decision" > isn't really appropriate, I suggest. > > The idea of "decision" seems wrapped up with "free will", which has a pretty > tenuous relationship with physical reality. > > If what you mean is that probabilities of events are associated with the > actions that agents take, then of course this is true.
Yes, no free will thing. But probabilities can also be used as random numbers, as art feeding mathematical taste or as a tool of mistaken decision-taking-method cult. In this case they don't serve as a method for aligning thought processes of agent with reality. > The (extremely) speculative hypothesis I was proposing in my blog post > is that perhaps intelligent agents can take two kinds of actions -- those > that are lateral moves within a given multiverse, and those that pop out > of one multiverse into another (surfing through the Yverse to another > multiverse). > > One could then talk about conditional probabilities of agent actions ... > which seems unproblematic ... OK, but why can't they all be dumped in a single 'normal' multiverse? If traveling between them is accommodated by 'decisions', there is a finite number of them for any given time, so it shouldn't pose structural problems. Another question is that it might be useful to describe them as organized in a tree-like structure, according to navigation methods accessible to an agent. If you represent uncertainty by being in 'more-parent' multiverse, it expresses usual idea with unusual (and probably unnecessarily restricting) notation. -- Vladimir Nesov mailto:[EMAIL PROTECTED] ----- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=4007604&id_secret=90496476-c3b0fe
