Session: Applications To Signal Processing
Chair: Robert Bassett
Cluster: Optimization Applications (Communication, Energy, Health, ML, ...)
Talk 1: Stochastic Optimal Search with Signals
Speaker: Jefferson Huang
Abstract: We consider the problem of optimally searching for a target located at one of the nodes of a network. On each time step, the searcher receives a (possibly unreliable) signal indicating where the target is on the network. Starting with the case of searching on a line, we explore the structure of search policies that optimally make use of the signals.
Talk 2: Sparse signal reconstruction for over-dispersed low-photon count imaging
Speaker: Roummel Marcia
Abstract: Low-photon count imaging is typically modeled by Poisson statistics. This discrete probability distribution model assumes that the mean and variance of a signal are equal. In the presence of greater variability in a dataset than what is expected, the negative binomial distribution is a suitable over-dispersed alternative to the Poisson distribution. In this work, we present an optimization framework for reconstructing sparse signals in these over-dispersed low-count settings.
Talk 3: Decentralized Sensor Network Localization via Customized Proximal Splitting
Speaker: Peter Barkley
Abstract: We apply recent advances in the design of custom proximal splitting algorithms to build a decentralized algorithm for solving the node-based SDP relaxation of the sensor network localization problem using noisy distance data. We implement the algorithm using MPI. We then explore the performance of various algorithm design options, and compare the decentralized algorithm with other decentralized approaches, including a graph-based decentralized ADMM.
