This question is mostly directed to Markus, if he's listening, since
he has significant experience with neural network programming and Go
programming.
Given the current MCTS landscape, is there any room for neural
networks in computer Go? If so, how might they play a role?
___
>Yes i mean (as Nick did) enthalpy
Nick's formula is entropy. The enthalpy formula is very different.
Nick's information-gathering analogy is also apt: by collecting trials that
have greater entropy your search is collecting more information. This leads
directly to a theory of why dynamic komi mi
Le 05/02/2011 15:17, Stefan Kaitschick a écrit :
You really mean enthalpy?
I thought entropy of information was more applicable here than enthalpy
of formation. :-)
(i don't get your joke, but this due to my poor english understanding :) )
Yes i mean (as Nick did) enthalpy
Temperature of the
Am 05.02.2011 12:50, schrieb Alain Baeckeroot:
Le 04/02/2011 14:48, Stefan Kaitschick a écrit :
Am 04.02.2011 14:36, schrieb Robert Lupton the Good:
Actually that's the Entropy, S; Enthalpy is E - PV (energy corrected
for pressure work).
If you are interested in thermodynamic analogues, the
Le 04/02/2011 14:48, Stefan Kaitschick a écrit :
Am 04.02.2011 14:36, schrieb Robert Lupton the Good:
Actually that's the Entropy, S; Enthalpy is E - PV (energy corrected
for pressure work).
If you are interested in thermodynamic analogues, the Helmholtz free
energy (E - TS) might be an inter
Reminder - it's tomorrow.
The February 2011 KGS computer Go tournament will be on Sunday February
6th, starting at 08:00 UTC and ending at 16:00 UTC.
It will be a 16-round Swiss with 19x19 boards, 14 minutes each of main
time, and Canadian Overtime of 25 moves in 60 seconds. It will use
Chi
