Learnability and the Vapnik-Chervonenkis dimension
Citations
37 citations
Cites background or methods from "Learnability and the Vapnik-Chervon..."
...This corollary follows immediately from the characterization above and is based on Blumer et al. (1989). We apply narrowing in order to narrow the class to VC dimension v....
[...]
...Therefore, it is infinite and by the fundamental theorem of learnability (Blumer et al., 1989; Vapnik & Chervonenkis, 1971), the concept class H is not learnable....
[...]
...One of the early breakthroughs in statistical learning theory was the seminal work of Vapnik & Chervonenkis (1971) and the later work of Blumer et al. (1989), which characterized binary classification settings as learnable if and only if the VC dimension is finite....
[...]
...This corollary follows immediately from the characterization above and is based on Blumer et al. (1989)....
[...]
37 citations
37 citations
37 citations
Cites background from "Learnability and the Vapnik-Chervon..."
...[6] improved the original bound of O(d/ log(d/ )) proved by Haussler and Welzl [17] to O(d/ log(1/ ))) D ow nl oa de d 01 /3 0/ 16 to 1 34 ....
[...]
...[6](4)) states that there exists an -net of size O(d/ log 1/ ) for any range space with VC-dimension d....
[...]
36 citations
Additional excerpts
...Vapnik–Chervonenkis (VC) dimension[1]....
[...]
References
42,654 citations
40,020 citations
17,939 citations
14,948 citations
13,647 citations