Class separability estimation and incremental learning using boundary methods

Neurocomputing

Volumn 35, Issue 1-4, 2000, Pages 3-26

  Sancho, José Luis ^a   Pierson, William E ^b   Ulug, Batu ^b   Figueiras Vidal, Aníbal R ^a   Ahalt, Stanley C ^b  

^a UNIVERSIDAD CARLOS III DE MADRID   (Spain)

^b Ohio State University   (United States)

Author keywords

Baye's error estimation; Class separability estimation; Classification neural systems; Gradient algorithms; Sample selection strategies

Indexed keywords

ERROR ANALYSIS; LEARNING ALGORITHMS; MULTILAYER NEURAL NETWORKS;

CLASS SEPARABILITY ESTIMATION; ELLIPTIC BOUNDARY METHODS (EBM); RADIAL BASIS FUNCTIONS (RBF); SAMPLE SELECTION (SS) MECHANISM;

LEARNING SYSTEMS;

ALGORITHM; ANALYTICAL ERROR; ARTICLE; ARTIFICIAL NEURAL NETWORK; BAYES THEOREM; CLASSIFICATION; CONTROLLED STUDY; INFORMATION; LEARNING; MATHEMATICAL ANALYSIS; METHODOLOGY; NOISE; PRIORITY JOURNAL; PROBLEM SOLVING; SAMPLING; TRAINING;

EID: 0034331286     PISSN: 09252312     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0925-2312(00)00293-9     Document Type: Article

References (30)

1. On a measure of divergence between two statistical populations defined by their probability distributions
2. Learning without local minima in radial basis function networks
3. Optimal learning in artificial neural networks: A review of theoretical results
4. Neural networks for pattern recognition
5. Pedagogical pattern selection strategies
6. On information and distance measures, error bounds and feature selection
7. A measure of asymptotic efficiency for tests of a hypothesis based on the sum of observations
8. On the resolvability of normally distributed vector parameter estimates
9. A course in density estimation
10. Pattern classification and scene analysis
11. Neural networks learning and expert systems
12. On the problem of local minima in backpropagation
13. Comments on local minima free conditions in multilayer perceptrons
14. Neural networks: A comprehensive foundation
15. Introduction to the theory of neural computation
16. Feature extraction based on decision boundaries
17. Bayesian methods for adaptive models
18. On estimation of a probability density function and mode
19. Using boundary m'ethods for estimating class separability
20. Theoretical and complexity issues for feature set evaluation using boundary methods
21. Selecting concise training sets from clean data
22. Learning internal representations by error propagation
23. Boundary methods for distribution analysis
24. Decision, estimation, and classification: An introduction to pattern recognition and related topics
25. Semidefinite programming
26. A representation theorem and its application to spherically-invariant random processes
27. Focused incremental learning for improved generalization with reduced training sets
28. Accelerated learning by active example selection