Sample Compression, Learnability, and the Vapnik-Chervonenkis Dimension

Machine Learning

Volumn 21, Issue 3, 1995, Pages 269-304

  Floyd, Sally ^a   Warmuth, Manfred ^b  

^a LAWRENCE BERKELEY NATIONAL LABORATORY   (United States)

^b UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

pac learning; Sample compression; Vapnik Chervonenkis dimension

Indexed keywords

COMBINATORIAL MATHEMATICS; COMPUTATION THEORY; FUNCTIONS; LEARNING ALGORITHMS;

HYPOTHESIS; MAXIMUM AND MAXIMAL CLASSES; PAC-LEARNING; RECONSTRUCTION FUNCTION; SAMPLE COMPRESSION; VAPNIK-CHERVONENKIS DIMENSION;

LEARNING SYSTEMS;

EID: 0029521676     PISSN: 08856125     EISSN: 15730565     Source Type: Journal    
DOI: 10.1023/A:1022660318680     Document Type: Article

References (32)

1. Queries and concept learning
2. Occam's razor
3. Learnability and the Vapnik-Chervonenkis dimension
4. The power of self-directed learning
5. Epsilon-nets & simplex range queries
6. Learning nested differences of intersection-closed concept classes
7. Learning integer lattices
8. Learning when irrelevant attributes abound: A new linear-threshold algorithm
9. Predicting {0, 1}-functions on randomly drawn points
10. Computational limitations on learning from examples
11. Inferring decision trees using the minimum description length principle
12. Stochastic complexity and modeling
13. On the density of families of sets
14. The strength of weak learnability
15. A theory of the learnable
16. On the uniform convergence of relative frequencies of events to their probabilities