Terrence Kwasha wrote:
Intelligence can be thought of as a classifier, as classification is a precursor to prediction which is a precursor to decision making. You have to classify something as an instance of what you have seen before, so you can predict it will do something within the range of what you have seen that class of things do before.
All known methods of algorithmic classification suffer from the problem of overfit. Example: All training photos of a classroom have a blackboard, so a room with a whiteboard can't be a classroom. This is wrong. Really the general concept of <Medium of communication> is a more useful for classification, but known algorithmic methods are not designed to create such generalizations. Humans are.