We present an approach to semi-supervised learning based on an exponential family characterization. Our approach generalizes previous work on coupled priors for hybrid generative/discriminative models. Our model is more flexible and natural than previous approaches. Experimental results on several data sets show that our approach also performs better in practice.

Biomedical image segmentation plays a significant role in computer-aided diagnosis. However, existing CNN based methods rely heavily on massive manual annotations, which are very expensive and require huge human resources. In this work, we adopt coarse-to-fine strategy propose self-supervised correction learning paradigm for semi-supervised biomedical segmentation. Specifically, design dual-tas...

We study semi-supervised learning when the data consists of multiple intersecting manifolds. We give a finite sample analysis to quantify the potential gain of using unlabeled data in this multi-manifold setting. We then propose a semi-supervised learning algorithm that separates different manifolds into decision sets, and performs supervised learning within each set. Our algorithm involves a n...

For efficient diagnosis processes, the multitude of heterogeneous medical data requires seamless integration. In order to automatically align radiology reports and images based on the pathological anatomical entities they describe, a preceding sentence classification is necessary. However, the lexical resource used has to contain semantic information about the pathological classification of eac...

The Universum sample, which is defined as the sample that doesn’t belong to any of the classes the learning task concerns, has been proved to be helpful in both supervised and semi-supervised settings. The former works treat the Universum samples equally. Our research found that not all the Universum samples are helpful, and we propose a method to pick the informative ones, i.e., inbetween Univ...

In this paper, we propose to study the problem of identifying and classifying tweets into intent categories. For example, a tweet “I wanna buy a new car” indicates the user’s intent for buying a car. Identifying such intent tweets will have great commercial value among others. In particular, it is important that we can distinguish different types of intent tweets. We propose to classify intent ...

For many real-world application problems, the availability of data labels for supervised learning is rather limited. It is often the case that a limited number of labelled cases is accompanied by a larger number of unlabeled ones. This is the setting for semi-supervised learning, in which unsupervised approaches assist the supervised problem and viceversa. In this report, we outline some basic ...

This paper proposes and develops a new graph-based semi-supervised learning method. Different from previous graph-based methods that are based on discriminative models, our method is essentially a generative model in that the class conditional probabilities are estimated by graph propagation and the class priors are estimated by linear regression. Experimental results on various datasets show t...

We propose a general approach to modeling semi-supervised learning constraints on unlabeled data. Both traditional supervised classification tasks and many natural semisupervised learning heuristics can be approximated by specifying the desired outcome of walks through a graph of classifiers. We demonstrate the modeling capability of this approach in the task of relation extraction, and experim...