Abstract
Let ℋ and ℓ be sets of functions from domain X to ℝ. We say that ℋ validly generalises ℓ from approximate interpolation if and only if for each η > 0 and ε δ ∈ (0, 1) there is m0(η, ε δ) such that for any function t ∈ ℓ and any probability distribution ℘ on X, if m ≥ m0 then with ℘m-probability at least 1 - δ, a sample X = (x1, x2,...,xm) ∈ Xm satisfies ∀h ∈ ℋ ιh(xi) - t(xi)ι <η, (1 ≤ i ≤m) ⇒ ℘({x :ιh(x) - t(x)ι ≥ η}) < e. We find conditions that are necessary and sufficient for ℋ to validly generalise ℓ from approximate interpolation, and we obtain bounds on the sample length m0(η ε δ) in terms of various parameters describing the expressive power of ℋ.
Original language | English |
---|---|
Pages (from-to) | 191-214 |
Number of pages | 24 |
Journal | Combinatorics Probability and Computing |
Volume | 5 |
Issue number | 3 |
DOIs | |
State | Published - 1996 |
ASJC Scopus subject areas
- Theoretical Computer Science
- Statistics and Probability
- Computational Theory and Mathematics
- Applied Mathematics