X hits on this document

143 views

0 shares

0 downloads

0 comments

14 / 40

Learnability Models

Three learnability definitions

Identification in the limit t (t is finite) gt = gt+1=…= g is correct

Finite identification h(i1, i2, …, it) is a decision function returning 0/1; g = G(i1, i2, …, it) iff h(i1, i2, …, it)=1

Fixed-time identification g = G(i1, i2, …, i) where is a constant

Document info
Document views143
Page views143
Page last viewedMon Jan 23 13:36:59 UTC 2017
Pages40
Paragraphs326
Words1871

Comments