On 3/1/07, Matt Mahoney <[EMAIL PROTECTED]> wrote:
The proof requires the assumption that the environment be computable.
Just to clarify what is meant here. What is required is that the environment has a computable distribution (in fact something even weaker than this but that's getting into too many technical details...) For example, a fair coin has a simple uniform distribution, and this distribution function is clearly computable. Thus, even though the sequence of results from these coin flips is not computable (with probability 1), the distribution is computable and thus this environment falls within the AIXI/Solomonoff framework. In this sense, everything that we know of in the universe is computable, including quantum mechanics. I don't know the details of the latter, but Hutter has a PhD in quantum physics and assures me that this is indeed the case. Shane ----- 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/?list_id=11983