[HN Gopher] A Theory of Universal Learning [pdf]
       ___________________________________________________________________
        
       A Theory of Universal Learning [pdf]
        
       Author : emre
       Score  : 40 points
       Date   : 2020-11-09 19:11 UTC (3 hours ago)
        
 (HTM) web link (web.math.princeton.edu)
 (TXT) w3m dump (web.math.princeton.edu)
        
       | bra-ket wrote:
       | Calling this 'universal learning' is a stretch (and in HN speak
       | 'click-bait'). The paper only talks about a particular subclass
       | of learning algorithms in supervised learning domain within PAC
       | framework.
       | 
       | Specially they talk about learning algorithms that minimize some
       | error function from training examples. That's not how learning
       | happens in living organisms. Therefore calling it 'universal' is
       | an unwarranted generalization, kind of like 'theory of
       | everything'.
       | 
       | A more appropriate name would be 'Distribution-dependent
       | Supervised PAC Learning' or something along those lines. It's a
       | solid work which addresses a particular niche of a particular
       | theory.
        
       ___________________________________________________________________
       (page generated 2020-11-09 23:02 UTC)