>Legg proved there is no such thing as a simple, universal learner. So we can stop looking for one.
With all due respect to everyone involved this kind of comprehensive sweeping statement is both narrow minded and counter productive. >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 is nonsensical, especially with regard to AGI, just the temporal aspect of reality invalidates the premise within any meaningful context. Predication is not a learned quality, it’s a natural phenomena generated by physical fields/ laws. (Pressure/ inertia/ gravity/ time/ etc) Take the bow/ pressure wave of a ship as an analogy. The bow wave is an accurate prediction generated by the medium (water) of where the ship will be in the future. The properties of the bow wave are also affected by the state of the water, waves or wakes from other passing ships. In the big scheme of things this predictive schema is a very simple system yet it’s impossible to recreate the exact sequence or indeed recreate the qualities of the predication post happening. This video shows a similar effect in a neural schema, the resultant pressure wave or inertia of the GTP thought pattern is generated by the preceding thought patterns, they are unique to this experience and would have totally different properties post happening… your program can’t run a copy and create a predictor because the prediction was partially temporally based… and the time has passed. https://www.youtube.com/watch?v=I1Dyj5hgvtc Language and mathematics are constructs created by an intelligent system; they are not an insight into how the intelligent system functions. :) ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T1ff21f8b11c8c9ae-Mf1c94ca843df1698b8964378 Delivery options: https://agi.topicbox.com/groups/agi/subscription
