Merging Uniform Inductive Learners
The fundamental learning model considered here is identification of recursive functions in the limit as introduced by Gold , but the concept is investigated on a meta-level. A set of classes of recursive functions is uniformly learnable under an inference criterion I, if there is a single learner, which synthesizes a learner for each of these classes from a corresponding description of the class. The particular question discussed here is how unions of uniformly learnable sets of such classes can still be identified uniformly. Especially unions of classes leading to strong separations of inference criteria in the uniform model are considered. The main result is that for any pair (I, I′) of different inference criteria considered here there exists a fixed set of descriptions of learning problems from I, such that its union with any uniformly I-learnable collection is uniformly I′-learnable, but no longer uniformly I-learnable.
KeywordsInference Criterion Recursive Function Inductive Inference Hypothesis Space Uniform Model
Unable to display preview. Download preview PDF.
- 2.Barzdin, J., Two Theorems on the Limiting Synthesis of Functions, Theory of Algorithms and Programs, Latvian State University, Riga 210, 82–88 (1974) (in Russian).Google Scholar
- 3.Barzdin, J., Inductive Inference of Automata, Functions and Programs, in Proc. International Congress of Math., Vancouver, 455–460 (1974).Google Scholar
- 12.Rogers, H., Theory of Recursive Functions and Effective Computability, MIT Press, Cambridge, Massachusetts (1987).Google Scholar
- 13.Wiehagen, R., Zeugmann, T., Learning and Consistency, in K.P. Jantke and S. Lange (Eds.): Algorithmic Learning for Knowledge-Based Systems, LNAI 961, 1–24, Springer-Verlag (1995).Google Scholar
- 14.Zilles, S., On the Comparison of Inductive Inference Criteria for Uniform Learning of Finite Classes, in N. Abe, R. Khardon, and T. Zeugmann (Eds.): ALT 2001, LNAI 2225, 251–266, Springer-Verlag (2001).Google Scholar