On the exponential value of labeled samples

Pattern Recognition Letters

Volumn 16, Issue 1, 1995, Pages 105-111

  Castelli, Vittorio ^a   Cover, Thomas M ^a  

^a STANFORD UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

DECISION THEORY; ERRORS; LEARNING SYSTEMS; NUMERICAL METHODS; OPTIMIZATION; PROBABILITY; RANDOM PROCESSES;

BAYES ERROR PROBABILITY; BAYES RISK; BERNOULLI RANDOM VARIABLES; DECISION REGIONS; EXPONENTIAL VALUE; EXPONENTS; LABELED SAMPLES; MIXTURES OF ELEMENTS; OPTIMAL ERROR PROBABILITY; UNLABELED SAMPLES;

PATTERN RECOGNITION;

EID: 0029195475     PISSN: 01678655     EISSN: None     Source Type: Journal    
DOI: 10.1016/0167-8655(94)00074-D     Document Type: Article

References (18)

1. Automatic pattern recognition: a study of the probability of error
2. Texture analysis via unsupervised and supervised learning
3. Dimensionality and sample size considerations in pattern recognition practice
4. Estimating the linear discriminant function from initial samples containing a small number of unclassified observations
5. Updating the discriminant function on the basis of unclassified data
6. Normal discrimination with unclassified observations
7. Combined use of supervised and unsupervised learning for dynamic security assessment
8. Small sample size effects in statistical pattern recognition: recommendations for practitioners
9. On the asymptotic improvement of supervised learning by utilizing additional unlabeled samples; normal mixture density case
10. On the mixtures of distributions
11. Identifiability of finite mixtures
12. Nonlinear mapping with minimal supervised learning