Optimal Unsupervised Learning in Feedforward Neural Networks

We investigate the properties of feedforward neural networks trained with Hebbian learning algorithms. A new unsupervised algorithm is proposed which produces statistically uncorrelated outputs. The algorithm causes the weights of the network to converge to the eigenvectors of the input correlation with largest eigenvalues. The algorithm is closely related to the technique of Self-supervised Backpropagation, as well as other algorithms for unsupervised learning. Applications of the algorithm to texture processing, image coding, and stereo depth edge detection are given. We show that the algorithm can lead to the development of filters qualitatively similar to those found in primate visual cortex.

A Detailed Look at Scale and Translation Invariance in a Hierarchical Neural Model of Visual Object Recognition

The HMAX model has recently been proposed by Riesenhuber & Poggio as a hierarchical model of position- and size-invariant object recognition in visual cortex. It has also turned out to model successfully a number of other properties of the ventral visual stream (the visual pathway thought to be crucial for object recognition in cortex), and particularly of (view-tuned) neurons in macaque inferotemporal cortex, the brain area at the top of the ventral stream. The original modeling study only used ``paperclip'' stimuli, as in the corresponding physiology experiment, and did not explore systematically how model units' invariance properties depended on model parameters. In this study, we aimed at a deeper understanding of the inner workings of HMAX and its performance for various parameter settings and ``natural'' stimulus classes. We examined HMAX responses for different stimulus sizes and positions systematically and found a dependence of model units' responses on stimulus position for which a quantitative description is offered. Interestingly, we find that scale invariance properties of hierarchical neural models are not independent of stimulus class, as opposed to translation invariance, even though both are affine transformations within the image plane.

Neural Networks

We present an overview of current research on artificial neural networks, emphasizing a statistical perspective. We view neural networks as parameterized graphs that make probabilistic assumptions about data, and view learning algorithms as methods for finding parameter values that look probable in the light of the data. We discuss basic issues in representation and learning, and treat some of the practical issues that arise in fitting networks to data. We also discuss links between neural networks and the general formalism of graphical models.

Forecasting Global Temperature Variations by Neural Networks

Global temperature variations between 1861 and 1984 are forecast usingsregularization networks, multilayer perceptrons and linearsautoregression. The regularization network, optimized by stochasticsgradient descent associated with colored noise, gives the bestsforecasts. For all the models, prediction errors noticeably increasesafter 1965. These results are consistent with the hypothesis that thesclimate dynamics is characterized by low-dimensional chaos and thatsthe it may have changed at some point after 1965, which is alsosconsistent with the recent idea of climate change.s

Biologically Plausible Neural Circuits for Realization of Maximum Operations

Object recognition in the visual cortex is based on a hierarchical architecture, in which specialized brain regions along the ventral pathway extract object features of increasing levels of complexity, accompanied by greater invariance in stimulus size, position, and orientation. Recent theoretical studies postulate a non-linear pooling function, such as the maximum (MAX) operation could be fundamental in achieving such invariance. In this paper, we are concerned with neurally plausible mechanisms that may be involved in realizing the MAX operation. Four canonical circuits are proposed, each based on neural mechanisms that have been previously discussed in the context of cortical processing. Through simulations and mathematical analysis, we examine the relative performance and robustness of these mechanisms. We derive experimentally verifiable predictions for each circuit and discuss their respective physiological considerations.

Biologically Plausible Neural Model for the Recognition of Biological Motion and Actions

The visual recognition of complex movements and actions is crucial for communication and survival in many species. Remarkable sensitivity and robustness of biological motion perception have been demonstrated in psychophysical experiments. In recent years, neurons and cortical areas involved in action recognition have been identified in neurophysiological and imaging studies. However, the detailed neural mechanisms that underlie the recognition of such complex movement patterns remain largely unknown. This paper reviews the experimental results and summarizes them in terms of a biologically plausible neural model. The model is based on the key assumption that action recognition is based on learned prototypical patterns and exploits information from the ventral and the dorsal pathway. The model makes specific predictions that motivate new experiments.

Exact Solution of the Nonlinear Dynamics of Recurrent Neural Mechanisms for Direction Selectivity

Different theoretical models have tried to investigate the feasibility of recurrent neural mechanisms for achieving direction selectivity in the visual cortex. The mathematical analysis of such models has been restricted so far to the case of purely linear networks. We present an exact analytical solution of the nonlinear dynamics of a class of direction selective recurrent neural models with threshold nonlinearity. Our mathematical analysis shows that such networks have form-stable stimulus-locked traveling pulse solutions that are appropriate for modeling the responses of direction selective cortical neurons. Our analysis shows also that the stability of such solutions can break down giving raise to a different class of solutions ("lurching activity waves") that are characterized by a specific spatio-temporal periodicity. These solutions cannot arise in models for direction selectivity with purely linear spatio-temporal filtering.

Memòria d’una indústria desapareguda: la bòbila d’en Soler

En aquest estudi pretenc revalorar el patrimoni cultural, natural i industrial de Vilafant. Tot i que les directrius del meu treball, es centren de manera especial en el patrimoni industrial, ja que en aquests moments l’Ajuntament de Vilafant, té especial interès en potenciar l’estudi de l’activitat rajolera, que va assolir el seu màxim esplendor amb la construcció de la bòbila d’en Soler l’any 1880

Muntatge d’una memòria en el sistema microprocessat Olorim

Aquest projecte parteix d'un projecte anterior realitzat per un company d'escola, en el qual es pretenia muntar un sistema per obtenir un diagnòstic dels pacients que pateixen bruxisme. El sistema que aquest company va muntar constava de dos subsistemes: el sistema de captura, encarregat de capturar el senyal mitjançant sensors i pretractar el senyal i el sistema de processament de dades, encarregat de rebre les dades provinents del sistema de captura mitjançant una ràdio sintonitzada a la freqüència 432,95MHz, que després s'envien al convertidor A/D de l'Olorim i s'emmagatzemen a la memòria interna de l'Olorim. Aquest projecte pretén millorar l'apartat de capacitat per a les dades i oferir major portabilitat mitjançant una targeta SD. Per dur a terme aquesta millora es recullen les dades emmagatzemades a la memòria interna del sistema microprocessat i s’emmagatzemen en una memòria SD. Les dades s'emmagatzemen a la targeta SD dins un fitxer creat prèviament amb l'ordinador, el qual ha de ser el primer fitxer que es crea a la targeta, ja que ha d'estar en sectors consecutius. En aquest fitxer s'aniran emmagatzemant les dades que ens proporcioni el sistema de captura en format RAW

Efficient learning of reactive robot behaviors with a Neural-Q_learning approach

The purpose of this paper is to propose a Neural-Q_learning approach designed for online learning of simple and reactive robot behaviors. In this approach, the Q_function is generalized by a multi-layer neural network allowing the use of continuous states and actions. The algorithm uses a database of the most recent learning samples to accelerate and guarantee the convergence. Each Neural-Q_learning function represents an independent, reactive and adaptive behavior which maps sensorial states to robot control actions. A group of these behaviors constitutes a reactive control scheme designed to fulfill simple missions. The paper centers on the description of the Neural-Q_learning based behaviors showing their performance with an underwater robot in a target following task. Real experiments demonstrate the convergence and stability of the learning system, pointing out its suitability for online robot learning. Advantages and limitations are discussed

Semi-online neural-Q_leaming for real-time robot learning

Reinforcement learning (RL) is a very suitable technique for robot learning, as it can learn in unknown environments and in real-time computation. The main difficulties in adapting classic RL algorithms to robotic systems are the generalization problem and the correct observation of the Markovian state. This paper attempts to solve the generalization problem by proposing the semi-online neural-Q_learning algorithm (SONQL). The algorithm uses the classic Q_learning technique with two modifications. First, a neural network (NN) approximates the Q_function allowing the use of continuous states and actions. Second, a database of the most representative learning samples accelerates and stabilizes the convergence. The term semi-online is referred to the fact that the algorithm uses the current but also past learning samples. However, the algorithm is able to learn in real-time while the robot is interacting with the environment. The paper shows simulated results with the "mountain-car" benchmark and, also, real results with an underwater robot in a target following behavior

Memòria de les activitats del museu arqueològic d'Eivissa i Formentera durant l'any 2003.'Memoria de las actividades del museo arqueológico de Ibiza y Formentera durante el año 2003'.

Se expone un breve repaso sobre las actuaciones llevadas a cabo por el Museo en Ibiza y Formentera el año 2003. Se especifican diversos campos de los cuales destacamos las actividades culturales realizadas. A continuación se nombran todas ellas: organización de actividades y participación en la segunda feria de la Ciencia y la tecnología realizada en Palma de Mallorca, realización de un seminario de arqueología de la granada nazarí, participación en la organización de actividades para la celebración del Día Internacional de los Museos, organización de jornadas de arqueología fenicio-púnica y, finalmente, visitas guiadas del museo para escolares.

Inca : una memòria del curs escolar 1917-1918. 'Inca : una memoria del curso escolar 1917-1918'.

Se presenta la coyuntura educativa del curso escolar 1917-1918 en la población de Inca (Mallorca) a partir del análisis de una memoria realizada por D. Antoni Ferrer Fanals, maestro de la escuela pública número 1. Esta memoria forma parte de un legajo no catalogado, es un pliego oficial con tres caras escritas a mano, y consta de tres puntos. En primer lugar presenta una relación de los trabajos y actividades que se realizaron durante el año; en segundo lugar, relaciona los resultados obtenidos (alumnado matriculado, asistencia, etc); finalmente, los obstáculos que han dificultado las tareas escolares, entre los que destaca la notoria indiferencia de los padres y el poco interés que sienten por la instrucción y educación de sus hijos.