The connectionist network has learned the correct pattern for a concept when
A) the connection weights add up to exactly +1.00.
B) the output pattern matches the initial input pattern and this symmetry becomes "locked" into the system.
C) the error signals are reduced to nearly none and the correct properties are assigned.
D) the output unit response is greater than zero and the input signal has to compensate.

Question

Quizplus · Accepted Answer

The correct pattern for a concept is learned when the error signals are reduced to nearly none and the correct properties are assigned. This means that the network is able to make accurate predictions and classify new inputs correctly. The other options do not necessarily indicate that the network has learned the correct pattern for a concept.

The Connectionist Network Has Learned the Correct Pattern for a Concept