Learnability and the Vapnik-Chervonenkis dimension
Citations
77 citations
77 citations
77 citations
75 citations
75 citations
Cites background or result from "Learnability and the Vapnik-Chervon..."
...Third, the results of [Blumer et aI, 1987] imply that we can only expect to learn a class of functions F if F has finite V-C dimension....
[...]
...This would suffice to assure that the hypothesis half space so generated would (with confidence 1 -0) have error less than €, as is seen from [Blumer et aI, 1987, Theorem A3....
[...]
...In particular, if F has Vapnik-Chervonenkis (V-C) dimension l1 d, then it has been proved[Blumer et al, 1987] that all A needs to do to be a valid learning algorithm is to call MO(f, 8, d) = max(~logj, Sfdlog1f3) examples and to find in polynomial time a function 9 E F which correctly classifies…...
[...]
...Thus, for example, it is simple to show that the class H of half spaces is Valiant learnable[Blumer et aI, 1987]....
[...]
...First, although the results of [Blumer et al., 1987] tell us we can gather enough information for learning in polynomial time, they say nothing about when we can actually find an algorithm A which learns in polynomial time....
[...]
References
42,654 citations
40,020 citations
17,939 citations
14,948 citations
13,647 citations