Overembedding method for modeling nonstationary systems.

We propose a general overembedding method for modeling and prediction of nonstationary systems. It basically enlarges the standard time-delay-embedding space by inclusion of the (unknown) slow driving signal, which is estimated simultaneously with the intrinsic stationary dynamics. Our method can be implemented with any modeling tool. Using, in particular, artificial neural networks, its application to both synthetic and real-world time series shows that it is highly efficient, leading to much more accurate results and longer prediction horizons than other existing overembedding methods in the literature.

Nonstationary time-series analysis: accurate reconstruction of driving forces.

We propose a simple method for the accurate reconstruction of slowly changing external forces acting on nonlinear dynamical systems. The method traces the evolution of the external force by locally linearizing the map dependency with the shifting parameter. Application of our algorithm to synthetic data corresponding to discrete models of evolving ecosystems shows an accuracy that outperforms those of previous methods in the literature. In addition, an application to the real-world sunspot time series recovers recently reported changes in solar activity during the last century.

A late-stopping method for optimal aggregation of neural networks.

Ensembles of artificial neural networks have been used in the last years as classification/regression machines, showing improved generalization capabilities that outperform those of single networks. However, it has been recognized that for aggregation to be effective the individual networks must be as accurate and diverse as possible. An important problem is, then, how to tune the aggregate members in order to have an optimal compromise between these two conflicting conditions. We propose here a simple method for constructing regression/classification ensembles of neural networks that leads to overtrained aggregate members with an adequate balance between accuracy and diversity. The algorithm is favorably tested against other methods recently proposed in the literature, producing an improvement in performance on the standard statistical databases used as benchmarks. In addition, and as a concrete application, we apply our method to the sunspot time series and predict the remainder of the current cycle 23 of solar activity.

Cancer de mama: diagnostico asistido por computadoras / Breast cancer: computational assisted diagnosis

Las microcalcificaciones agrupadas en mamografias son un signo temprano de una lesion maligna en casi la mitad de los canceres de mama. La caracterizacion de las lesiones benignas y malignas representa un problema sumamente complejo aun para un radiologo con experiencia. Esto se refleja en el alto porcentaje de biopsias innecesarias que se llevan a cabo. La deteccion y clasificacion automatica pueden ayudar al radiologo en la evaluacion de las microcalcificaciones agrupadas. Desarrollamos un esquema de diagnostico asistido por computadora, utilizando hardware de PC, que automaticamente detecta, segmenta y clasifica las microcalcificaciones. El proceso de deteccion esta basado en una combinacion de filtros Gaussiano y morfologico. Dos clasificadores estadisticos fueron usados para la diferenciacion entre los casos malignos y los benignos, los cuales se basan en un conjunto seleccionado de caracteristicas cuantitativas. Para una tasa de falsos positivos del 10 por ciento, una tasa de verdaderos positivos del 87 por ciento fue alcanzada con el clasificador que utiliza el metodo de los K vecinos mas proximos

Cancer de mama: diagnostico asistido por computadoras / Breast cancer: computational assisted diagnosis

Las microcalcificaciones agrupadas en mamografias son un signo temprano de una lesion maligna en casi la mitad de los canceres de mama. La caracterizacion de las lesiones benignas y malignas representa un problema sumamente complejo aun para un radiologo con experiencia. Esto se refleja en el alto porcentaje de biopsias innecesarias que se llevan a cabo. La deteccion y clasificacion automatica pueden ayudar al radiologo en la evaluacion de las microcalcificaciones agrupadas. Desarrollamos un esquema de diagnostico asistido por computadora, utilizando hardware de PC, que automaticamente detecta, segmenta y clasifica las microcalcificaciones. El proceso de deteccion esta basado en una combinacion de filtros Gaussiano y morfologico. Dos clasificadores estadisticos fueron usados para la diferenciacion entre los casos malignos y los benignos, los cuales se basan en un conjunto seleccionado de caracteristicas cuantitativas. Para una tasa de falsos positivos del 10 por ciento, una tasa de verdaderos positivos del 87 por ciento fue alcanzada con el clasificador que utiliza el metodo de los K vecinos mas proximos (AU)

Learning persistent dynamics with neural networks.

Persistence is one of the most common characteristics of real-world time series. In this work we investigate the process of learning persistent dynamics by neural networks. We show that for chaotic times series the network can get stuck for long training periods in a trivial minimum of the error function related to the long-term autocorrelation in the series. Remarkably, in these cases the transition to the trained phase is quite abrupt. For noisy dynamics the training process is smooth. We also consider the effectiveness of two of the most frequently used decorrelation methods in avoiding the problems related to persistence. Copyright 1997 Elsevier Science Ltd.

Persistence of nonoptimal strategies.

Metastable configurations in open computational systems with local minima in their optimality functions are shown to be very long lived, which makes them effectively stable. When rare transitions to the global optimum do occur, they happen extremely fast, in analogy to models of punctuated evolution in biology. These results are obtained by introducing a thermodynamic-like formalism that allows for a simple analysis of nonlinear game dynamics in computational ecosystems.