7.4. Two-Class Probabilistic DBNNs
A probabilistic decision-based neural network (PDBNN) (e.g., described in [208, 210-212]) is a probabilistic variant of its predecessor—the decision-based neural network (DBNN) [192]. This section describes the network structure and learning rules of the PDBNN for binary classification problems. The extension to multiclass pattern classification is presented in Section 7.5.
Consider an the example of binary classification (hypothesis testing) problems based on an independent and identical distributed (i.i.d.) observation sequence:
Hypothesis Ω0: xn is defined by the probability distribution p(x|Ω0) for n= 1,2, ..., N.
Hypothesis Ω1: xn is defined by the probability distribution p(x|Ω1).
A conventional approach ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access