Sequential Maximum Margin Classifiers for Partially Labeled Data

TL;DR

This paper introduces a sequential maximum margin classifier framework that updates incrementally with new data, leveraging the Maximum Entropy Discrimination principle to handle large feature sets and unlabeled data.

Contribution

It presents a novel sequential learning approach for maximum margin classifiers using the Maximum Entropy Discrimination principle, accommodating unlabeled data and kernel representations.

Findings

Performs comparably or better than non-sequential classifiers

Effective with large feature sets and unlabeled data

Validated on simulated and real datasets

Abstract

In many real-world applications, data is not collected as one batch, but sequentially over time, and often it is not possible or desirable to wait until the data is completely gathered before analyzing it. Thus, we propose a framework to sequentially update a maximum margin classifier by taking advantage of the Maximum Entropy Discrimination principle. Our maximum margin classifier allows for a kernel representation to represent large numbers of features and can also be regularized with respect to a smooth sub-manifold, allowing it to incorporate unlabeled observations. We compare the performance of our classifier to its non-sequential equivalents in both simulated and real datasets.