Boltzmann Machines vs Back-Propagation networks
I'm wondering, are there any circumastances within game AI where you would want to use a Boltzmann Machine (A kind of noisy version of the Hopfield Net, with a hidden layer) over Back-Propagation networks?
Back-Propagation nets seem to be much faster, both in training and execution. I'm just wondering whether anyone sees any advantage in using the Boltzmann Machine instead? ...
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement