Next: Guidelines and Rules
Up: Learning in Feed-Forward
Previous: Conjugate Gradients
Even though the target values t in a classifying problem are binary,
the output units in an MLP will not take values that are exactly
1 or 0. However, one can interpret these outputs in a very useful way;
they correspond to Bayesian a posteriori probabilities
 provided that:
This very important result enables the network outputs to be further
in a controlled way. In the case of function mapping the output error
can be estimated with standard methods based on distances to cluster
centers in the training data.
- The training is accurate.
- The outputs are of 1-of-M-type (the task is coded
such that only one output unit is ``on'' at a time).
- A mean square, cross entropy or Kullback error function is used.
- Training data points are selected with the correct a priori
System PRIVILEGED Account
Fri Feb 24 11:28:59 MET 1995