We develop a computationally efficient framework for finding compact and highly accurate hidden-variable models via entropy minimization. The main results are: 1) An entropic prior that favors small, unambiguous, maximally structured models. 2) A priorbalancing manipulation of Bayes’ rule that allows one to gradually introduce or remove constraints in the course of iterative reestimation. #1 an...

A Wireless Sensor Network (WSN) is a collection of small sensor nodes with sensing, computation and communication capabilities. The sensor nodes are randomly deployed in distributed environment. The main goal of this paper is to form clusters of sensor nodes and to perform Data Aggregation to collect the data by maintaining the energy efficiency so that the network lifetime can be increased. Ne...

We examine the Extended Poisson Model proposed by [2] for encoding and decoding probability distributions in populations of neurons. We derive the equations for two MAP Estimation methods: ExpectationMaximization and Softmax, to reconstruct an input distribution based on the observed firing activity of the neurons. We analyze the performance of these decoding methods in extracting information r...

We investigate the problem of learning discrete, undirected graphical models in a differentially private way. We show that the approach of releasing noisy sufficient statistics using the Laplace mechanism achieves a good trade-off between privacy, utility, and practicality. A naive learning algorithm that uses the noisy sufficient statistics “as is” outperforms general-purpose differentially pr...

Abstract. Allele-specific expression (ASE) refers to the differential abundance of the allelic copies of a transcript. Direct RNA sequencing (RNA-Seq) can provide quantitative estimates of ASE for genes with transcribed polymorphisms. However, estimating ASE is challenging due to ambiguities in read alignment. Current approaches do not account for the hierarchy of multiple read alignments to ge...

In linear mixed models, the assumption of normally distributed random effects is often inappropriate and unnecessarily restrictive. The proposed approximate Dirichlet process mixture assumes a hierarchical Gaussian mixture that is based on the truncated version of the stick breaking presentation of the Dirichlet process. In addition to the weakening of distributional assumptions, the specificat...

The scalability problem in data mining involves the development of methods for handling large databases with limited computational resources. In this paper, we present a two-phase scalable model-based clustering framework: First, a large data set is summed up into sub-clusters; Then, clusters are directly generated from the summary statistics of sub-clusters by a specifically designed Expectati...

Expectation-maximization algorithms, such as those implemented in GIZA++ pervade the field of unsupervised word alignment. However, these algorithms have a problem of over-fitting, leading to “garbage collector effects,” where rare words tend to be erroneously aligned to untranslated words. This paper proposes a leave-one-out expectationmaximization algorithm for unsupervised word alignment to ...

Linking individuals across historical datasets relies on information such as name and age that is both non-unique and prone to enumeration and transcription errors. These errors make it impossible to find the correct match with certainty. We suggest a fully automated method for linking historical datasets that enables researchers to create samples that minimize type I (false positives) and type...

Either directly or indirectly, the lexicon for a natural language specifies complementation frames or valences for open-class words such as verbs and nouns. Constructing a lexicon of complementation frames for large vocabularies constitutes a challenge of scale, with the further complication that frame usage, like vocabulary, varies with genre and undergoes ongoing innovation in a living langua...