Having multiple perceptrons can actually solve the XOR problem satisfactorily: this is because each perceptron can partition off a linear part of the space itself, and they can then combine their results.

A. true - this works always, and these multiple perceptrons learn to classify even complex problems

B. false - perceptrons are mathematically incapable of solving linearly inseparable functions, no matter what you do

C. true - perceptrons can do this but are unable to learn to do it - they have to be explicitly hand-coded

D. false - just having a single perceptron is enough

Answer: Option C