Inducing a classification function from a set of examples in the form of labeled instances is a standard problem in supervised machine learning. In this paper, we are concerned with ambiguous label classification (ALC), an extension of this setting in which several candidate labels may be assigned

that filters noisy continuous labeled examples, whose computational cost is O(N·logN+NA 2) for N examples and A attributes. Besides, it is shown experimentally to be better than the embedded algorithms of the state-of-the art of the Machine Learning Systems. 1

We propose a family of learning algorithms based on a new form of regularization that allows us to exploit the geometry of the marginal distribution. We focus on a semi-supervised framework that incorporates labeled and unlabeled data in a general-purpose learner. Some transductive graph learning

The Problem: Learning to recognize objects from very few labeled training examples, but large numbers of unlabeled examples. Motivation: Statistical object recognition techniques require large training sets to achieve good performance. It is often difficult and expensive to collect many labeled

We consider the problem of using a large unlabeled sample to boost performance of a learning algorithm when only a small set of labeled examples is available. In particular, we consider a setting in which the description of each example can be partitioned into two distinct views, motivated

Abstract not found

twenty years have brought many changes but the song remains the same. The title of the book indicates that as we develop the theory, we will focus our attention on examples. Hoping that the book would be a useful reference for people who apply probability in their work, we have tried to emphasize

We present a system for identifying the semantic relationships, or semantic roles, filled by constituents of a sentence within a semantic frame. Various lexical and syntactic features are derived from parse trees and used to derive statistical classifiers from hand-annotated training data. 1

, and demonstrate that most of the steps are polynomial, typically bounded by O (N 2 ). Next we demonstrate some examples of its operation taken from our cognitive simulation studies and work in machine learning. Finally, we compare SME to other analogy programs and discuss several areas for future work. This paper

of labeling many samples. We apply these semisupervised learning methods to a data set of cars and background images, attempting to separate the two classes. Some of the algorithms obtain very high classification accuracy and can be used towards a car-detection system.