Luis Sucar

Followers

Following

Co-authors

Public Views

Ernesto Castaneda

American University

Gildardo Sánchez

Tecnológico de Monterrey

Marc Champagne

Kwantlen Polytechnic University

Armando Marques-Guedes

UNL - New University of Lisbon

Aleksandar Shurbevski

Kyoto University

Viacheslav Kuleshov

Stockholm University

Nikolaus Egel

Universität Münster

Güzin Özdağoğlu

Dokuz Eylül University

Carla Rahn Phillips

University of Minnesota

Panos Vassiliadis

University of Ioannina

Interests

Uploads

Papers by Luis Sucar

Clonación conductista de habilidades de control de humano a máquina: un enfoque bayesiano

by Luis Sucar and Julio A Freyre-Gonzalez

El objetivo de esta investigación es el de evaluar técnicas bayesianas de clasificación aplicadas... more El objetivo de esta investigación es el de evaluar técnicas bayesianas de clasificación aplicadas a la clonación conductista de operadores humanos. Para esto, se desarrolló y estudió un modelo probabilístico bayesiano, con discretización uniforme, para la clonación de habilidades de control de humano a máquina, basándose en el modelo de clonación conductista (behavioural cloning) establecido por Michie et al. Como crisol de prueba del modelo propuesto se utilizó el problema del carro con el péndulo invertido. Se obtuvieron buenos resultados en el control a pesar de la simplicidad del método utilizado. Además, se observó que la discretización y el retardo son parámetros muy importantes para la inducción de los clones. Por otra parte, se concluyó que al incrementar el numero de ejemplos, se gana robustez en los clones inducidos. Por último, se observó que el porcentaje de aciertos no es un buen indicador de que tan bien un clon inducido pueda controlar el sistema dinámico.

Download

Indoor Mobile Robots Using Ultrasonic and Laser Range Sensors

Automatic image annotation using semi-supervised generative modeling

Pattern Recognition, 2015

ABSTRACT Automatic image annotation consists on automatically labeling images, or image regions, ... more ABSTRACT Automatic image annotation consists on automatically labeling images, or image regions, with a pre-defined set of keywords, which are regarded as descriptors of the high-level semantics of the image. In supervised learning, a set of previously annotated images is required to train a classifier. Annotating a large quantity of images by hand is a tedious and time consuming process; so an alternative approach is to label manually a small subset of images, using the other ones under a semi-supervised approach. In this paper, a new semi-supervised ensemble of classifiers, called WSA, for automatic image annotation is proposed. WSA uses naive Bayes as its base classifier. A set of these is combined in a cascade based on the AdaBoost technique. However, when training the ensemble of Bayesian classifiers, it also considers the unlabeled images on each stage. These are annotated based on the classifier from the previous stage, and then used to train the next classifier. The unlabeled instances are weighted according to a confidence measure based on their predicted probability value; while the labeled instances are weighted according to the classifier error, as in standard AdaBoost. WSA has been evaluated with benchmark data sets, and 2 sets of images, with promising results.

An Exploration Approach for Indoor Mobile Robots Reducing Odometric Errors

Lecture Notes in Computer Science, 2002

Abstract. To learn a map of an environment a mobile robot has to ex-plore its workspace. This pap... more

Learning Probabilistic Grid-Based Maps for Indoor Mobile Robots Using Ultrasonic and Laser Range Sensors

Lecture Notes in Computer Science, 2000

Abstract. A new method for learning probabilistic grid-based maps of the environment of a mobile ... more

Solving the Global Localization Problem for Indoor Mobile Robots

Lecture Notes in Computer Science, 2003

ABSTRACT Global localization is the problem of determining the position of a robot under global u... more ABSTRACT Global localization is the problem of determining the position of a robot under global uncertainty. This problem can be divided in two phases: 1) from the sensor data (or sensor view), determine clusters of hypotheses where the robot can be; and 2) devise a strategy by which the robot can correctly eliminate all but the right location. In the second phase, previous approaches consider an ideal robot, a robot with a perfect odometer, to predict robot movements. This paper introduces a non deterministic prediction approach based on a Markov localization that include an uncertainty model for the movements of the robot. The non deterministic model can help to solve situations where a deterministic or ideal model fails. Hypotheses are clustered and a greedy search algorithm determines the robot movements to reduce the number of clusters of hypotheses. This approach is tested using a simulated mobile robot with promising results.

Exploration and Navigation for Mobile Robots With Perceptual Limitations

International Journal of Advanced Robotic Systems, 2006

Abstract: To learn a map of an environment a mobile robot has to explore its workspace using its ... more

Integrating Bayesian Networks with Logic Programs for Music

Integrating bayesian networks with logic programs

A Bi-directional approach for learning Bayesian networks and its application to ozone prediction

Recognition of Continuous Activities

Page 1. Recognition of Continuous Activities Rocío Díaz de León, L. Enrique Sucar ITESM-Campus Cu... more

Towards a Region-Level Automatic Image Annotation Benchmark

Abstract Automatic image annotation at region-level consists of the task of assigning labels to r... more Abstract Automatic image annotation at region-level consists of the task of assigning labels to regions within segmented images. This is a very important task for the development and improvement of image retrieval methods. Many image annotation methods have been proposed so far, reporting relatively good results on this task. However the lack of benchmark collections have caused that region-level methods are often evaluated using image-level collections. On the other hand, a few region-level collections are available, ...

A Flexible Service Robot Named Markovito

Word Co-occurrence and Markov Random Fields for Improving Automatic Image

Abstract In this paper a novel approach for improving automatic image annotation methods is propo... more Abstract In this paper a novel approach for improving automatic image annotation methods is proposed. The approach is based on the fact that accuracy of current image annotation methods is low if we look at the most confident label only. Instead, accuracy is improved if we look for the correct label within the set of the top− k candidate labels. We take advantage of this fact and propose a Markov random field (MRF) based on word co-occurrence information for the improvement of annotation systems. Through the MRF structure we ...

A General Framework for Developing Service Robots

We present a general framework for developing ser- vice robots that can help people in daily acti... more We present a general framework for developing ser- vice robots that can help people in daily activi- ties. This framework is based on a general, dis- tributed architecture which integrates several compo- nents; (i) coordinator, (ii) navigator, (iii) perception, and (iv) human-robot interface. The coordinator uses a decision-theoretic approach to select the appropri- ate action according to the current state.

Gesture Therapy - A Low-Cost Vision-Based System for Rehabilitation after Stroke

by Gildardo Azcarate and Luis Sucar

México debe de invertir en Robots de Servicio

An energy-based model for feature selection

Abstract In this paper we propose an energy-based model (EBM) for selecting subsets of features t... more Abstract In this paper we propose an energy-based model (EBM) for selecting subsets of features that are both causally and predictively relevant for classification tasks. The proposed method is tested in the causality challenge, a competition that promotes research on strengthen feature selection by taking into account causal information of features. Under the proposed approach, an energy value is assigned to every configuration of features and the problem is reduced to that of finding the configuration that minimizes an energy ...

Ambientes de Aprendizaje en Robots de Servicio

Special Track on Uncertain Reasoning