Download:
|
by Susanne Kaufmann, Efim Kinber, Martin Kummer
http://werner.ira.uka.de/~kaufmann/CKKK95ea.ps
Add To MetaCart
Abstract:
Abstract. Investigated is algorithmic learning, in the limit, of correct programs for recursive functions f from both input/output examples of f and several interesting varieties of approximate additional (algorithmic) information about f. Specifically considered, as such approximate additional information about f, are Rose's frequency computations for f and several natural generalizations from the literature, each generalization involving programs for restricted trees of recursive functions which have f as a branch. Considered as the types of trees are those with bounded variation, finite width, and finite rank. For the case of learning final correct programs for recursive functions, EX-learning, where the additional information involves frequency computations, an insightful and interestingly complex combinatorial characterization of learning power is presented as a function of the frequency parameters. Pitt has previously characterized the learning power of probabilistic machines in terms of the size of equally powerful teams of (deterministic) machines. For EX-learning (as well as for BC-learning, where a final sequence of correct programs is learned), for the cases of providing the types of additional information considered in this paper, the optimal machine team size is determined such that the entire class of recursive functions is learnable. 1
Citations
|
536
|
Learnability and the Vapnik-Chervonenkis Dimension
– Blumer, Ehrenfeucht, et al.
- 1989
|
|
365
|
Recursively Enumerable Sets and Degrees
– Soare
- 1987
|
|
229
|
Classical Recursion Theory
– Odifreddi
- 1989
|
|
147
|
Comparison of identification criteria for machine inductive inference, Theoretical Computer Science 25
– Case, Smith
- 1983
|
|
56
|
The power of pluralism for automatic program synthesis
– SMITH
- 1982
|
|
48
|
Approximable sets
– Beigel, Kummer, et al.
- 1995
|
|
48
|
Robot Planning
– McDermott
- 1992
|
|
42
|
Terse, superterse, and verbose sets
– Beigel, Gasarch, et al.
- 1993
|
|
41
|
Probability and plurality for aggregations of learning machines. Information and Computation
– PITT, SMITH
|
|
35
|
Learning via queries
– Gasarch, Smith
- 1992
|
|
33
|
Polynomial terse sets
– Amir, Gasarch
- 1988
|
|
27
|
A proof of Beigel's cardinality conjecture
– Kummer
- 1992
|
|
27
|
Probabilistic inductive inference
– Pitt
- 1989
|
|
23
|
A Cardinality Version of Beigel's Nonspeedup Theorem
– OWINGS
- 1989
|
|
13
|
An extended notion of computability
– Rose
- 1960
|
|
12
|
Frequency computation and the cardinality theorem
– Harizanov, Kummer, et al.
- 1992
|
|
10
|
Learning with higher order additional information
– Baliga, Case
- 1994
|
|
10
|
Inductive inference with additional information
– Freivalds, Wiehagen
- 1979
|
|
10
|
On frequency computation of functions. Algebra i Logika, 2:25--32
– Trakhtenbrot
- 1963
|
|
9
|
Training sequences
– Angluin, Gasarch, et al.
- 1989
|
|
7
|
Inclusion problems in parallel learning and games
– Kummer, Stephan
- 1994
|
|
6
|
Quantifying the amount of verboseness. To appear in: Information and Computation. (A preliminary version appeared in
– Beigel, Kummer, et al.
- 1992
|
|
6
|
On frequency calculations of general recursive predicates
– Kinber
- 1972
|
|
5
|
On a quantitative notion of uniformity
– Kaufmann, Kummer
- 1995
|
|
4
|
On (m,n)-computable sets
– Degtev
- 1981
|
|
4
|
Frequency-computable functions and frequency-enumerable sets. Candidate Dissertation
– Kinber
- 1975
|
