On the limits of efficient teachability

Information Processing Letters

Volumn 79, Issue 6, 2001, Pages 267-272

  Servedio, Rocco A ^a  

^a HARVARD UNIVERSITY   (United States)

Author keywords

Computational learning theory; Machine learning; Teaching dimension

Indexed keywords

BOOLEAN FUNCTIONS; LEARNING ALGORITHMS; LEARNING SYSTEMS; POLYNOMIALS; PROBABILITY DISTRIBUTIONS;

TEACHING SEQUENCES;

TEACHING;

EID: 0035975706     PISSN: 00200190     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0020-0190(01)00137-5     Document Type: Article

References (20)

1. On specifying boolean functions using labelled examples
2. On learning a union of halfspaces
3. Polynomial time algorithms for learning neural nets
4. Learning with unreliable boundary queries
5. Learning an intersection of a constant number of halfspaces over a uniform distribution
6. Training a 3-node neural net is NP-complete
7. Inductive inference - An abstract approach
8. On the complexity of teaching
9. Teaching a smarter learner
10. Generalized teaching dimensions and the query complexity of learning
11. How many queries are needed to learn?
12. A computational model of teaching
13. PAC learning intersections of halfspaces with membership queries
14. Algorithms and lower bounds for on-line learning of geometrical concepts
15. A model of interactive teaching
16. Learning with a helpful teacher
17. Teachability in computational learning
18. Short monotone formulae for the majority function
19. A random sampling based algorithm for learning the intersection of halfspaces