"Suppose you have a simple learner that can predict any computable sequence of symbols with some probability at least as good as random guessing. Then I can create a simple sequence that your predictor will get wrong 100% of the time. My program runs a copy of your program and outputs something different from your guess."
This kind of program is an example of narrow AGI, and the application of the theory as a proof that a universal learner is impossible is irrelevant. It does not apply to all forms of knowledge, in particular, the kind of knowledge that we work with all of the time. There is no basis that the prediction made by a program like this could be absolutely right all of the time. The misunderstanding that a 'predictor' is the same as absolute knowledge that is always right has no basis in the world that might be known from common sense. This is not a proof that a universal learner is impossible because the foundation of knowledge is not the striving for perfect knowledge of the future. ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T1ff21f8b11c8c9ae-M686945359db215b9199c28a4 Delivery options: https://agi.topicbox.com/groups/agi/subscription
