> If you compute P(AllPossibleModels|AllSensoryInput) you end up never learning anything.
...whoa. That makes complete sense.
So you're saying that there must be some form of meta-rationality that gives cues to our attempts at Bayesian reasoning, directing those attempts how to make selections from each set (the set of all possible models and the set of all sensory inputs) in order to produce results that constitute actual learning.
And you're suggesting that in animals and humans at least, the feedback loop of our embodied experience is at least some part of that meta-rationality.
...whoa. That makes complete sense.
So you're saying that there must be some form of meta-rationality that gives cues to our attempts at Bayesian reasoning, directing those attempts how to make selections from each set (the set of all possible models and the set of all sensory inputs) in order to produce results that constitute actual learning.
And you're suggesting that in animals and humans at least, the feedback loop of our embodied experience is at least some part of that meta-rationality.
That's an incredible one-liner.