We introduce a neural-network architecture, termed the constrained recurrent sparse autoencoder (CRsAE), that solves convolutional dictionary learning problems, thus establishing link between and neural networks. Specifically, we leverage interpretation of alternating-minimization algorithm for as an approximate Expectation-Maximization to develop autoencoders enable simultaneous training regul...

This paper considers parameter estimations in Lomax distribution under progressive type-II censoring with random removals, assuming that the number of units removed at each failure time has a binomial distribution. The maximum likelihood estimators (MLEs) are derived using the expectation-maximization (EM) algorithm. The Bayes estimates of the parameters are obtained using both the squared erro...

We develop an expectation-maximization algorithm with local adaptivity for image segmentation and classification. The key idea of our approach is to combine global statistics extracted from the Gaussian mixture model or other proper statistical models with local statistics and geometrical information, such as local probability distribution, orientation, and anisotropy. The combined information ...

In three-dimensional (3-D) fluorescence microscopy, a series of two-dimensional (2-D) images is collected at different focal settings through the specimen. Each image in this series contains the in-focus plane plus contributions from out-of-focus structures that blur the image. Furthermore, as the series is collected the fluorescent dye in the specimen fades over time in response to the total e...

Planning for multiple agents under uncertainty is often based on decentralized partially observable Markov decision processes (DecPOMDPs), but current methods must de-emphasize long-term effects of actions by a discount factor. In tasks like wireless networking, agents are evaluated by average performance over time, both short and longterm effects of actions are crucial, and discounting based s...

Image restoration keeping sharp edges is achieved by bilateral filter. In this paper, an approach to improve edges for the filter is presented. The proposed algorithm relies on clustering by Expectation Maximization that produced clusters of intensive values. A stage is followed where standard deviation of Gaussian filters for scales of the spatial and intensity are adjusted by features of the ...

The ‘expectation–conditional maximization either’ (ECME) algorithm has proven to be an effective way of accelerating the expectation–maximization algorithm for many problems. Recognizing the limitation of using prefixed acceleration subspaces in the ECME algorithm, we propose a dynamic ECME (DECME) algorithm which allows the acceleration subspaces to be chosen dynamically. The simplest DECME im...

Expectation Maximization (EM) algorithm is a parameter estimation method from incomplete observations. In this paper, an implementation of this method to the calibration of HKS spectrometer at Jefferson Lab is described. We show that the application of EM method is able to calibrate the spectrometer properly in the presence of high background noise, while the traditional nonlinear Least Square ...