V. Vapnik and A. Chervonenkis. Ordered risk minimisation I. Automation and Remote Control, Vol. 35 pp.1226--1235, 1974 Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
On Design of Optimal Nonlinear Kernel Potential Function for.. - Hu, Li, Liang (2003) (Correct)
....function. For linear potential, regions are half spaces defined by hyperplanes, and the optimal hyperplane (w, b) is the one with maximal distance to the origin. This is related to the novelty detection problem and single class support vector machine studied in statistical learning theory [36, 39, 47]. In our case, any non protein points will need to be detected as outliers from the protein distribution characterized by cN cD . Among all linear functions derived from the same set of native proteins and decoys, the optimal weight vector w is likely to have the least amount of mislabellings. ....
.... by obtaining the parameters #D and #N from solving the following Lagrange dual form of quadratic programming problem: i#N#D # i i,j#N#D y i y j # i # j e subject to 0 # # i # C where C is a regularizing constant that limits the influence of each misclassified conformation [36 39, 47], and y i = 1 if i is a native protein, and y i = 1 if i is a decoy. These parameters lead to optimal classification of unseen test set proteins against decoys [36 39, 47] Modeling Sequence Design Potential. For protein sequence design, we assume that native sequence is more stable on the ....
[Article contains additional citation context not shown here]
Vapnik V, Chervonenkis A. A note on one class of perceptrons. Automation and Remote Control 1964;25.
Adapting Codes and Embeddings for Polychotomies - Rätsch, Smola, Mika (2003) (Correct)
....be related to a large margin classifier, beginning with binary classification. 2. 1 Large Margins Dichotomies Here a large margin classifier is defined as a mapping f : X R with the property that yf(x) or more specifically yn f(xn ) with (xn ; yn ) 2 Z, where is some positive constant [16]. Since such a positive margin may not always be achievable, one typically maximizes a penalized version of the maximum margin, such as f ] where yn f(xn ) 1 n ; n 0; n = 1; N and f 2 F: 1) Here f ] is a regularization term, 0 is a regularization constant and F ....
V.N. Vapnik and A.Y. Chervonenkis. A note on one class of perceptrons. Automation and Remote Control, 25, 1964.
Statistical Learning and Kernel Methods - Schölkopf (2000) (Correct)
....a hyperplane learning algorithm that can be performed in a dot product space (such as the feature space that we introduced previously) As described in the previous section, to design learning 5 algorithms, one needs to come up with a class of functions whose capacity can be computed. 32] and [30] considered the class of hyperplanes (w Delta x) b = 0 w 2 R ; b 2 R; 20) corresponding to decision functions f(x) sgn ( w Delta x) b) 21) and proposed a learning algorithm for separable problems, termed the Generalized Portrait, for constructing f from empirical data. It is based ....
V. Vapnik and A. Chervonenkis. A note on one class of perceptrons. Automation and Remote Control, 25, 1964. 26
Extracting Phonetic Knowledge from Learning Systems.. - Damper, Gunn, Gore (1999) (1 citation) (Correct)
....and velar) As for the perceptron experiments, the networks were trained using the same 100 patterns: 50 repetitions of responses to the 0 ms VOT stimuli and 50 repetitions of responses to the 80 ms VOT stimuli. A straightforward SVM was used with an architecture equivalent to a perceptron [13, 46, 47], with a hard limiting (signum function) threshold unit on the output (equation 1) There was no additional capacity control. The auditory data are easily separated with a linear hyperplane in the 192 dimensional feature space; as such, no additional capacity control to tolerate errors in ....
V. N. Vapnik and A. J. Chervonenkis. A note on the class of perceptrons. Automation and Remote Control, 25:103--109, 1964.
On the Flexibility of Theoretical Models - For Pattern Recognition (Correct)
No context found.
V. Vapnik and A. Chervonenkis. Ordered risk minimisation I. Automation and Remote Control, Vol. 35 pp.1226--1235, 1974
Adapting Codes and Embeddings for Polychotomies - Rätsch, Smola, Mika (Correct)
No context found.
V.N. Vapnik and A.Y. Chervonenkis. A note on one class of perceptrons. Automation and Remote Control, 25, 1964.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC