On Fri, Nov 09, 2007 at 03:40:19PM +0100, Shane Legg wrote:
[...]
!
I haven't finished reading the thing, but I did notice some typos.
Page 8: defn 1.3.2 has a missing \mu; it should say
"... has the additional property \mu(\Omega)..."
and next sentence is also missing a \mu:
should say "...then we call the tuple (\Omega, D, \mu)..."
bottom of page, should say
"... P \ {0} is isomorphic to A^{\omega}"
instead of
"... P \ {0} is isomorphic to A^*"
Page 10, should define the new notation not used previously:
\mu(xy|x) == \frac {\mu(xy)} {\mu (x)}
Page 10 .. Wikipedia seems to want to call this
"Kullback-Leibler divergence" not just plain "Kullback divergence"
Page 11 .. section 1.6, Fourth sentence of first paragraph,
a psi is missing a backslash \psi (so its spelled out, instead of a
symbol being printed.)
Section 1.4 seems to have a general self-inconsistency: sometimes
you are prepending, and sometimes appending to a string. When dealing
with infinite strings, you can only prepend/append to the start of the
string, not to the end (well, OK, you can append to the end of an
infinite string, but that launches you into transfinite math,
which I suspect is not at all the intent here). Since P is defined by
prepending, switching to appended notation for the rest of the section
confuses the issue. Fixing this would als allow your defin's and
theorems to work on A^\omega, and not just A^*.
Nice paper, though. I hope I will be pleasently entertained when I get
to section 2.
--linas
-----
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=8660244&id_secret=63537143-cf2914
