Local Learning Algorithms
Léon Bottou, Vladimir N. Vapnik · Neural Computation · 1992
Very rarely are training data evenly distributed in the input space. Local learning algorithms attempt to locally adjust the capacity of the training system to the properties of the training set in each area of the input space. The family of local learning algorithms contains known methods, like the k-nearest neighbors method (kNN) or the radial basis function networks (RBF), as well as new algorithms. A single analysis models some aspects of these algorithms. In particular, it suggests that neither kNN or RBF, nor nonlocal classifiers, achieve the best compromise between locality and capacity. A careful control of these parameters in a simple local learning algorithm has provided a performance breakthrough for an optical character recognition problem. Both the error rate and the rejection performance have been significantly improved.
Cited by 542 works. doi:10.1162/neco.1992.4.6.888
Authors
References
- Handwritten Digit Recognition with a Back-Propagation Network
- Probability Inequalities for sums of Bounded Random Variables
- Fast Learning in Networks of Locally-Tuned Processing Units
- Adaptive Mixtures of Local Experts
- Radial Basis Functions, Multi-Variable Functional Interpolation and Adaptive Networks
- Probability Inequalities for Sums of Bounded Random Variables
- Structural Risk Minimization for Character Recognition
- Similarity Metric Learning for a Variable-Kernel Classifier
- Principles of Risk Minimization for Learning Theory
- Transforming Neural-Net Output Levels to Probability Distributions