Author Archives: admin

In this talk we are interested in distributed algorithms for solving separable optimization problems. Many problems in engineering can be formulated as separable optimization problems, i.e., minimizing the sum of P functions subject to the intersection of P sets. Our … Read More

High-dimensional datasets are increasingly common in learning problems, in many different domains, such as text categorization, genomics, econometrics, and computer vision. The excessive number of features carries the problem of memory usage in order to represent and deal with these … Read More

The patterns in collections of real world objects are often not based on a limited set of isolated properties such as features. Instead, the totality of their appearance constitutes the basis of the human recognition of patterns. Structural pattern recognition … Read More

In this presentation we develop a dynamic continuous solution to the clustering problem of data characterized by a mixture of K distributions, where K is given a priori. The proposed solution resorts to game theory tools, in particular mean field … Read More

Speech enhancement techniques aim to recover the original clean signal underlying corrupted speech. Such techniques typically operate in the short-time Fourier transform (STFT) domain where phenomena like additivity of background noises, interfering speakers and echoes are easier to model. By … Read More

How to detect and classify credit card fraud? What is fraud for a person, say, buying expensive jewelry, spending on online casinos or simply spending a lot of money, is just the regular day-to-day for another. To make matters worse, … Read More

Linear models have enjoyed great success in structured prediction in NLP. While a lot of progress has been made on efficient training with several loss functions, the problem of endowing learners with a mechanism for feature selection is still unsolved. … Read More

The problem of separating synchronous sources (SSS) is a case of blind source separation (BSS) where independence of the sources is not satisfied. In SSS, the sources are assumed to be complex-valued, and different sources are phase-locked, which means that … Read More

We possess a wealth of prior knowledge about most prediction problems, and particularly so for many of the fundamental tasks in natural language processing. Unfortunately, it is often difficult to make use of this type of information during learning, as … Read More