Tuesday, February 17, 2009
The Recurrent Temporal Restricted Boltzmann Machine
This paper is succedent to the previous motion caption paper, which does not include much details about the model (at least I don't know anything about that dynamic biases). Anyway, I find more details in this paper. This dynamic bias is actually a cross term of ht and ht-1. So given ht-1, it is functioning as the bias term.
The model proposed in this paper is a bit strange. I will read something about recurrent networks first. From the point of view of optimization, it uses BP instead of CD learning (exact vs stochastic approximation).
Anyway, I will return to this paper later. It is really weird to make a temporal model with RBM compared with HMM, CRF.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment