accedaCRIShttps://accedacris.ulpgc.es/jspuiThe accedaCRIS digital repository system captures, stores, indexes, preserves, and distributes digital research material.Tue, 13 Aug 2024 17:42:00 GMT2024-08-13T17:42:00Z5051FPGA implementation of neurocomputational models: comparison between standard back-propagation and C-Mantec constructive algorithmhttp://hdl.handle.net/10553/114748Title: FPGA implementation of neurocomputational models: comparison between standard back-propagation and C-Mantec constructive algorithm
Authors: Ortega Zamorano, Francisco; Jerez, José M.; Juárez, Gustavo E.; Franco, Leonardo
Abstract: Recent advances in FPGA technology have permitted the implementation of neurocomputational models, making them an interesting alternative to standard PCs in order to speed up the computations involved taking advantage of the intrinsic FPGA parallelism. In this work, we analyse and compare the FPGA implementation of two neural network learning algorithms: the standard and well known Back-Propagation algorithm and C-Mantec, a constructive neural network algorithm that generates compact one hidden layer architectures with good predictive capabilities. One of the main differences between both algorithms is the fact that while Back-Propagation needs a predefined architecture, C-Mantec constructs its network while learning the input patterns. Several aspects of the FPGA implementation of both algorithms are analyzed, focusing in features like logic and memory resources needed, transfer function implementation, computation time, etc. The advantages and disadvantages of both methods in relationship to their hardware implementations are discussed.
Sun, 01 Jan 2017 00:00:00 GMThttp://hdl.handle.net/10553/1147482017-01-01T00:00:00ZPiecewise polynomial activation functions for feedforward neural networkshttp://hdl.handle.net/10553/114745Title: Piecewise polynomial activation functions for feedforward neural networks
Authors: López-Rubio, Ezequiel; Ortega Zamorano, Francisco; Domínguez, Enrique; Muñoz-Pérez, José
Abstract: Since the origins of artificial neural network research, many models of feedforward networks have been proposed. This paper presents an algorithm which adapts the shape of the activation function to the training data, so that it is learned along with the connection weights. The activation function is interpreted as a piecewise polynomial approximation to the distribution function of the argument of the activation function. An online learning procedure is given, and it is formally proved that it makes the training error decrease or stay the same except for extreme cases. Moreover, the model is computationally simpler than standard feedforward networks, so that it is suitable for implementation on FPGAs and microcontrollers. However, our present proposal is limited to two-layer, one-output-neuron architectures due to the lack of differentiability of the learned activation functions with respect to the node locations. Experimental results are provided, which show the performance of the proposal algorithm for classification and regression applications.
Tue, 01 Jan 2019 00:00:00 GMThttp://hdl.handle.net/10553/1147452019-01-01T00:00:00ZExploratory data analysis and foreground detection with the growing hierarchical neural foresthttp://hdl.handle.net/10553/114744Title: Exploratory data analysis and foreground detection with the growing hierarchical neural forest
Authors: Palomo, Esteban J.; López-Rubio, Ezequiel; Ortega Zamorano, Francisco; Benítez-Rochel, Rafaela
Abstract: In this paper, a new self-organizing artificial neural network called growing hierarchical neural forest (GHNF) is proposed. The GHNF is a hierarchical model based on the growing neural forest, which is a tree-based model that learns a set of trees (forest) instead of a general graph so that the forest can grow in size. This way, the GHNF faces three important limitations regarding the self-organizing map: fixed size, fixed topology, and lack of hierarchical representation for input data. Hence, the GHNF is especially amenable to datasets containing clusters where each cluster has a hierarchical structure since each tree of the GHNF forest can adapt to one of the clusters. Experimental results show the goodness of our proposal in terms of self-organization and clustering capabilities. In particular, it has been applied to text mining of tweets as a typical exploratory data analysis application, where a hierarchical representation of concepts present in tweets has been obtained. Moreover, it has been applied to foreground detection in video sequences, outperforming several methods specialized in foreground detection.
Wed, 01 Jan 2020 00:00:00 GMThttp://hdl.handle.net/10553/1147442020-01-01T00:00:00ZAuthentication of individuals using hand geometry biometrics: A neural network approachhttp://hdl.handle.net/10553/46162Title: Authentication of individuals using hand geometry biometrics: A neural network approach
Authors: Faundez-Zanuy, Marcos; Elizondo, David A.; Ferrer-Ballester, Miguel Ángel; Travieso-González, Carlos M.
Abstract: Biometric based systems for individual authentication are increasingly becoming indispensable for protecting life and property. They provide ways for uniquely and reliably authenticating people, and are difficult to counterfeit. Biometric based authenticity systems are currently used in governmental, commercial and public sectors. However, these systems can be expensive to put in place and often impose physical constraint to the users. This paper introduces an inexpensive, powerful and easy to use hand geometry based biometric person authentication system using neural networks. The proposed approach followed to construct this system consists of an acquisition device, a pre-processing stage, and a neural network based classifier. One of the novelties of this work comprises on the introduction of hand geometry’s related, position independent, feature extraction and identification which can be useful in problems related to image processing and pattern recognition. Another novelty of this research comprises on the use of error correction codes to enhance the level of performance of the neural network model. A dataset made of scanned images of the right hand of fifty different people was created for this study. Identification rates and Detection Cost Function (DCF) values obtained with the system were evaluated. Several strategies for coding the outputs of the neural networks were studied. Experimental results show that, when using Error Correction Output Codes (ECOC), up to 100% identification rates and 0% DCF can be obtained. For comparison purposes, results are also given for the Support Vector Machine method.
Mon, 01 Jan 2007 00:00:00 GMThttp://hdl.handle.net/10553/461622007-01-01T00:00:00ZAnt colony optimization inspired algorithm for 3D object segmentation into its constituent partshttp://hdl.handle.net/10553/44278Title: Ant colony optimization inspired algorithm for 3D object segmentation into its constituent parts
Authors: Arnay, Rafael; Acosta, Leopoldo; Sanchez-Medina, Javier
Abstract: This work focuses on the use of an Ant colony optimization (ACO) based approach to the problem of 3D object segmentation. The ACO metaheuristic uses a set of agents (artificial ants) to explore a search space. This kind of metaheuristic can be classified as a Natural computing non-deterministic technique, which is frequently used when the size of the search space makes the use of analytic mathematical tools unaffordable. The exploration is influenced by heuristic information, determined by each particular problem. Agents communicate with each other through the pheromone trails, which act as the common memory for the colony. In the approach presented, the agents start their exploration at the outer contour of an object. The final result is given after a certain number of generations, when the particular solutions of the agents converge to create the global paths followed by the colony. These paths coherently connect the object’s high curvature areas, facilitating the segmentation process. The advantage of this convergence mechanism is that it avoids the problem of over-segmentation by detecting regions based on the global structure of the object and not just on local information.
Thu, 01 Jan 2015 00:00:00 GMThttp://hdl.handle.net/10553/442782015-01-01T00:00:00Z