The key idea behind active learning is that a machine learning algorithm can achieve
greater accuracy with fewer labeled training instances if it is allowed to choose the training …

This paper addresses the repeated acquisition of labels for data items when the labeling is
imperfect. We examine the improvement (or lack thereof) in data quality via repeated …

In all these cases, labels can be obtained, but only at a significant cost to the end user. An
important observation is that all records are not equally important from the perspective of …

Feature selection helps reduce data acquisition costs in ML, but the standard approach is to
train models with static feature subsets. Here, we consider the dynamic feature selection …

Many real-life decision-making situations allow further relevant information to be acquired at
a specific cost, for example, in assessing the health status of a patient we may decide to take …

The diversity of an ensemble of classifiers is known to be an important factor in determining
its generalization error. We present a new method for generating ensembles, Decorate …

Deep generative models often perform poorly in real-world applications due to the
heterogeneity of natural data sets. Heterogeneity arises from data containing different types …

This paper addresses the repeated acquisition of labels for data items when the labeling is
imperfect. We examine the improvement (or lack thereof) in data quality via repeated …

We propose a Bayesian undirected graphical model for co-training, or more generally for
semi-supervised multi-view learning. This makes explicit the previously unstated …

Most of the empirical evaluations of active learning approaches in the literature have
focused on a single classifier and a single performance measure. We present an extensive …