The k-means algorithm is a popular clustering method used in many different fields of computer science, such as data mining, machine learning and information retrieval. However, the k-means algorithm is very likely to converge to some local optimum which is much worse than the desired global

scenarios. Both a robust formulation of the censoring problem, and a locally-optimum formulation of the censoring problem are considered. We nd conditions under which the robust censoring problem can be solved by designing for the least-favorable distributions from the uncertainty classes. For the locally-optimum

A locally optimum approach for estimating a nonrandom pa-rameter lying in some small neighborhood of a known nom-inal value is considered. Reference is made to a decentral-ized estimation problem in the context of wireless sensor net-works, and particular attention is paid to the design

A country’s suitability for entry into a currency union depends on a number of economic conditions. These include, inter alia, the intensity of trade with other potential members of the currency union, and the extent to which domestic business cycles are correlated with those of the other countries. But international trade patterns and international business cycle correlations are endogenous. This paper develops and investigates the relationship between the two phenomenon. Using thirty years of data for twenty industrialized countries, we uncover a strong and striking empirical finding: countries with closer trade links tend to have more tightly correlated business cycles. It follows that countries are more likely to satisfy the criteria for entry into a currency

Abstract: Ethernet is a branching broadcast communication system for carrying digital data packets among locally distributed computing stations. The packet transport mechanism provided by Ethernet has been used to build systems which can be viewed as either local computer networks or loosely

Abstract—We propose a new analytic method for comparing constant gain adaptive signal processing algorithms. Specifically, estimates of the convergence speed of the algorithms allow for the definition of a local measure of performance, called the efficacy, that can be theoretically evaluated

EM algorithm is a very popular iterationbased method to estimate the parameters of Gaussian Mixture Model from a large observation set. However, in most cases, EM algorithm is not guaranteed to converge to the global optimum. Instead, it stops at some local optimums, which can be much worse than

Based on local optimality, a novel sequential detection scheme is proposed in this paper. The performance of the proposed scheme is compared with that of the sequential probability ra-tio test (SPRT) and truncated SPRT (TSPRT). The proposed scheme is shown to have higher efficiency, lower

advantages include the flexibility of the procedures, as well as the ability to self-adapt the search for optimum solutions on the fly. As desktop computers increase in speed, the application of evolutionary algorithms will become routine. 1 Introduction Darwinian evolution is intrinsically a robust search

This is an informal unreviewed manuscript primarily intended for the exchange of information among NMC staff members.