Search results for "algorithm"

Experiments in Value Function Approximation with Sparse Support Vector Regression

2004

We present first experiments using Support Vector Regression as function approximator for an on-line, sarsa-like reinforcement learner. To overcome the batch nature of SVR two ideas are employed. The first is sparse greedy approximation: the data is projected onto the subspace spanned by only a small subset of the original data (in feature space). This subset can be built up in an on-line fashion. Second, we use the sparsified data to solve a reduced quadratic problem, where the number of variables is independent of the total number of training samples seen. The feasability of this approach is demonstrated on two common toy-problems.

2014

For locating inaccurate problem of the discrete localization criterion proposed by Demigny, a new criterion expression of “good localization” is proposed. Firstly, a discrete expression of good detection and good localization criterion of two dimension edge detection operator is employed, and then an experiment to measure optimal parameters of two dimension Canny's edge detection operator is introduced after. Moreover, a detailed performance comparison and analysis of two dimension optimal filter obtained via utilizing tensor product for one dimension optimal filter are provided which can prove that least square support vector regression (LS-SVR) is a smoothness filter and give the construc…

Optimal gossip algorithm for distributed consensus SVM training in wireless sensor networks

2009

In this paper, we consider the distributed training of a SVM using measurements collected by the nodes of aWireless Sensor Network in order to achieve global consensus with the minimum possible inter-node communications for data exchange. We derive a novel mathematical characterization for the optimal selection of partial information that neighboring sensors should exchange in order to achieve consensus in the network. We provide a selection function which ranks the training vectors in order of importance in the learning process. The amount of information exchange can vary, based on an appropriately chosen threshold value of this selection function, providing a desired trade-off between cla…

Multi-dimensional Function Approximation and Regression Estimation

2002

In this communication, we generalize the Support Vector Machines (SVM) for regression estimation and function approximation to multi-dimensional problems. We propose a multi-dimensional Support Vector Regressor (MSVR) that uses a cost function with a hyperspherical insensitive zone, capable of obtaining better predictions than using an SVM independently for each dimension. The resolution of the MSVR is achieved by an iterative procedure over the Karush-Kuhn-Tucker conditions. The proposed algorithm is illustrated by computers experiments.

Training label cleaning with ant colony optimization for classification of remote sensing imagery

2015

This paper presents an original approach for improving performances of the supervised classifiers in remote sensing imagery by proposing a technique to refine a given training set using Ant Colony Optimization (ACO). The new method called ACO-Training Label Cleaning (ACO-TLC) applies ACO model for selection of the significant training samples from a given set of labeled vectors in order to optimize the quality of a supervised classifier. This means to retain the most informative samples and to remove the uncertain or misclassified training samples, which lead to classification errors. As a result of the selection process, we can obtain a purified training set. The proposed model is implemen…

Calibration of the Norwegian motion laboratory using conformal geometric algebra

2017

This paper applies Conformal Geometric Algebra (CGA) as a tool for calibrating the robotic equipment found in the Norwegian Motion Laboratory. By using the inner product of CGA to measure the distance between a point and the surface of a plane/sphere, the least-squares method can be used to solve for the unknown parameters describing the plane/sphere in an efficient and intuitive way given n measured points. Positional data samples were acquired from using a high precision Laser tracker (FARO Xi), and the overall calibration error was found to be no more than 4.90mm, and the maximum standard deviation 3.25mm. In addition, the applied least-squares algorithm using CGA was twice as fast, when…

Ricci-flow based conformal mapping of the proximal femur to identify exercise loading effects.

2018

AbstractThe causal relationship between habitual loading and adaptive response in bone morphology is commonly explored by analysing the spatial distribution of mechanically relevant features. In this study, 3D distribution of features in the proximal femur of 91 female athletes (5 exercise loading groups representing habitual loading) is contrasted with 20 controls. A femur specific Ricci-flow based conformal mapping procedure was developed for establishing correspondence among the periosteal surfaces. The procedure leverages the invariance of the conformal mapping method to isometric shape differences to align surfaces in the 2D parametric domain, to produce dense correspondences across an…

GEPOL: An improved description of molecular surfaces. III. A new algorithm for the computation of a solvent-excluding surface

1994

To understand and calculate the interactions of a solute with a solvent, a good method of computing the molecular surface is needed. Three kinds of surfaces may be used: the van der Waals Surface, the Accessible Surface, and the Molecular Surface. The latter is redefined in this article as the Solvent-Excluding Surface. The new algorithm for computing the Solvent-Excluding Surface included in the GEPOL93 program is described. GEPOL93 follows the same concept as former versions of GEPOL but with a full new algorithm. Thus, it computes the Solvent-Excluding Surface by filling the spaces not accessible to the solvent with a set of new spheres. The computation is controlled by three parameters:…

Algorithms for the calculation of the view factors between human body and rectangular surfaces in parallelepiped environments

1992

Abstract The thermal comfort conditions for people in moderate thermal environments are subjected to spatial changes, depending on the radiative exchanges of the human body with the surrounding surfaces. Radiative thermal exchanges are notably accounted for by means of the “view factors” between a person in a given posture and the surface of the enclosure. These important parameters can be evaluated by means of a large set of graphs by Fanger, for rectangular surfaces. But the handling of graphs could lead to reading errors. In this paper a simple algorithm is presented, which is able to compute the required view factors. A validation, by means of a comparison against experimental data, is …

Morphological Analysis of Binary Scene in APR Integrated Environment

2009

This paper describes principles of binary scene [1] morphological analysis in script based application - APR (Analysis, Processing and Recognition). The aim of the method is to find object on the scene and then to describe theirs basic features like edges, neighbors and surface [2]. The algorithm construction gives benefits in terms speed as well as to computation costs, at the same time being capable of presenting number of attributes values for scene and each of the objects. There are also some practical algorithm applications showed.