Supervised Selective Combining Pattern Recognition Modalities and Its Application to Signature Verification by Fusing On-Line and Off-Line Kernels

Abstract

Abstract. We consider the problem of multi-modal pattern recognition under the assumption that a kernel-based approach is applicable within each particular modality. The Cartesian product of the linear spaces into which the respective kernels embed the output scales of single sensors is employed as an appropriate joint scale corresponding to the idea of combining modalities at the sensor level. This contrasts with the commonly adopted method of combining classifiers inferred from each specific modality. However, a significant risk in combining linear spaces is that of overfitting. To address this, we set out a stochastic method for encompassing modal-selectivity that is intrinsic to (that is to say, theoretically contiguous with) the selected kernel-based pattern-recognition approach. The principle of kernel selectivity supervision is then applied to the problem of signature verification by fusing several on-line and off-line kernels into a complete training and verification technique.

Supervised Selective Combination of Diverse Object-Representation Modalities for Regression Estimation

“…The greater the variety of different modalities, the broader the diversity of object properties which underlie the regression model (5). But if the number of kernels n is too large for the size N of the available training set, the overcomplicated model loses its generalization performance.…”

A Modified Neutral Point Method for Kernel-Based Fusion of Pattern-Recognition Modalities with Incomplete Data Sets

“…This fusion may be performed at the early or late stage. In the former case of early fusion [1,2], the growing overall dimensionality of the object representation with increasing the number of modalities can be reduced by incorporating some form of modalityselection within the final classification procedure [3,4], thereby eliminating the danger of over-fitting. Such modality-selectivity is correlated with the generalization performance of the training process, so that, if performed ideally, the recognition system user is free to include object-representation modalities without constraint.…”

Supervised Selective Kernel Fusion for Membrane Protein Prediction

“…For combination using several kernels we here utilize the generalized probabilistic formulation of the SVM, which was proposed in [18,20,21] as an instrument for making Bayesian decisions on the discriminant hyperplane…”

“…We refer to this property as "selectivity", because it defines an algorithm's ability to select kernels most useful to the classification task at hand. A generalized probabilistic approach for supervised selective kernel fusion was proposed by the authors in [21], [22] and includes, as particular cases, such familiar approaches as the classical SVM [6], Lasso SVM [23], Elastic Net SVM [24] and others.…”

“…Only a few methods are explicitly oriented towards elimination of this disadvantage and obtaining non-sparse decisions [17], [18] (more advanced versions utilize a supervised sparseness parameter [19], [20], [21], [22]). We refer to this property as "selectivity", because it defines an algorithm's ability to select kernels most useful to the classification task at hand.…”