CRIS

Permanent URI for this communityhttps://scripta.up.edu.mx/handle/20.500.12552/1

Browse

Settings

Has files: NoSubject: search.filters.subject.Metaheuristics

Search Results

Now showing 1 - 7 of 7
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Neural Architecture Search Using Trajectory Metaheuristics to Classify Coronary Stenosis
    (IEEE, 2024)
    Franco-Gaona, Erick
    ;
    Avila-Garcia, Maria-Susana
    ;
    Cruz-Aceves, Ivan
    ;
    Orocio-Garcia, Hiram-Efrain
    ;
    Escobedo-Gordillo, Andrés
    Coronary stenosis is a disease that claims millions of lives each year. Early detection of this condition is crucial for patient survival. Currently, physicians perform detection by x-ray angiograms, however, the variability of diagnoses and the difficulty of access to expertise has led to the need for automated, computer-assisted diagnosis. In this work explores the use of deep learning to classify stenosis or non-stenosis in angiogram images using convolutional neural networks from scratch. A methodology to fine-tuning network architectures automatically using metaheuristic optimization techniques is proposed, demonstrating superior performance to fine-tuning empirically and proposing a new architecture in the literature to classify coronary stenosis. The proposed architectures achieved 86.02% and 95.67% F1-score with simulated annealing and iterated local search techniques, respectively. ©The authors ©IEEE
      8
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    City Logistics
    (Springer, 2018)
    Barceló, Jaume
    ;
    Grzybowska, Hanna
    ;
    Orozco-Leyva, Arturo
    This chapter provides an introductory overview of city logistics systems, highlighting the specific characteristics that make them different from general logistics problems. It analyzes the types of decisions involved in managing city logistics applications, from strategic, tactical, and operational, and identifies the key models to address them. This analysis identifies types of problems, location, location routing, and variants of routing problems with time windows, all those with ad hoc formulations, derived from the constraints imposed by policy and operational regulations, technological conditions, or other specificities of urban scenarios, which result in variants of the classical models that, for its size and complexity, become a fertile field for metaheuristic approaches to define algorithms to solve the problems. Some of the more relevant cases are studied in this chapter, and guidelines for further and deeper insights on other cases are provided to the reader through a rich set of bibliographical references. ©Springer
    Scopus© Citations 2  13  1
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    MTGWA: A Multithreaded Gray Wolf Algorithm with Strategies Based on Simulated Annealing and Genetic Algorithms
    (2021)
    ;
    Murillo-Suarez, Alfonso
    ;
    García-Jacas, Cesar Raúl
    ;
    Guerrero-Valadez, Juan Manuel
    In this paper, we present an improvement of the Gray Wolf algorithm (GWO) based on a multi-threaded implementation of the original algorithm. The paper demonstrates how to combine the solutions obtained in each of the threads to achieve a final solution closer to the absolute minimum or even equal to it. To properly combine the solutions of each of the threads of execution, we use strategies based on simulated annealing and genetic algorithms. Also, we show the results obtained for twenty-nine functions: unimodal, multimodal, fixed dimension and composite functions. Experiments show that our proposed improves the results of the original algorithm. © Springer Nature
    Scopus© Citations 1  30  2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    A new swarm algorithm for global optimization of multimodal functions over multi-threading architecture hybridized with simulating annealing
    (2018)
    ;
    Murillo-Suarez, Alfonso
    This paper presents a new algorithm, PCLPSO, based on particle swarm optimization, which uses comprehensive learning particle swarm optimizer. Our algorithm executes C parallel CLPSO algorithms. We adopted as a criterion of completion a maximum value of evaluations of the objective function. During the execution of the CLPSO algorithms, when a certain evaluation value of the functions is reached, the best k are selected, and different initialization criteria are applied to continue the execution of the CLPSO algorithms: restarting the worst ones for the best solution or restores the worst ones to a random solution. For this restart, we use the Boltzmann criterion in a similar way as Simulating Annealing (SA) does. In this work, the experimental results obtained for the search of the minimum of 16 multimodal test functions such as Rosenbrock, Griewank, Rastrigin, Brannin, Schwefel, and others. Our algorithm proved to be more efficient than the traditional CLPSO in its experimental results, and the nonparametric Wilcoxon test confirmed this.
    Scopus© Citations 7  18  1
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Rain-Fall Optimization Algorithm with new parallel implementations
    (2018)
    Guerrero-Valadez, Juan Manuel
    ;
    Rainfall Optimization Algorithm (RFO) is a nature-inspired metaheuristic optimization algorithm. RFO mimics the movement of water drops generated during rainfall to optimize a function. The paper study new implementations for RFO to offer more reliable results. Moreover, it studies three restarting techniques that can be applied to the algorithm with multithreading. The different implementations for the RFO are benchmarked to test and verify the performance and accuracy of the solutions. The paper presents and compares the results using several multidimensional testing functions, as well as the visual behavior of the raindrops inside the benchmark functions. The results confirm that the movement of the artificial drops corresponds to the natural behavior of raindrops. The results also show the effectiveness of this behavior to minimize an optimization function and the advantages of parallel computing restarting techniques to improve the quality of the solutions.
    Scopus© Citations 1  4  2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    A new approach of the rain-fall optimization algorithm using parallelization
    (2020)
    Guerrero-Valadez, Juan Manuel
    ;
    This chapter introduces a new implementation of the Rain-Fall Optimization Algorithm (RFO) proposed by Kaboli, Sevbaraj, and Rahim in “Rain-Fall Optimization Algorithm. A Population-Based Algorithm for Solving Constrained Optimization Problems” by Kaboli et al. (J Comput Sci 19:31–42, 2017). RFO is a nature-inspired algorithm, which is based on the behavior of the water drops produced by a rainfall going down through a mountain to find the minimum values of specific functions. The algorithm was tested on four multidimensional benchmark functions: Ackley, Griewank, Rosenbrock, and Sphere functions. It was also tested in a four-dimensional function, the Kowalik function. The first step was to match the results of the rewritten algorithm with the results obtained by the original authors. Then the algorithm had to be modified to make some efficiency improvements and to get better results. The main modifications were a new equation to modify the step size for a function called explosion process and a parallel execution of the algorithm with two different restarting techniques: restart to the best and genetic restart to the best. © Springer Nature Switzerland AG 2020.
      50  1
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Towards the Distributed Wound Treatment Optimization Method for Training CNN Models: Analysis on the MNIST Dataset
    (IEEE, 2023)
    ;
    ;
    Convolutional neural network (CNN) is a prominent algorithm in Deep Learning methods. CNN architectures have been used successfully to solve various problems in image processing, for example, segmentation, classification, and enhancement task. However, automatic search for suitable architectures and training parameters remain an open area of research, where metaheuristic algorithms have been used to fine-tuning the hyperparameters and learning parameters. This work presents a bio-inspired distributed strategy based on Wound Treatment optimization (WTO) for training the learning parameters of a LenNet CNN model fast and accurate. The proposed method was evaluated over the popular benchmark dataset MNIST for handwritten digit recognition. Experimental results showed an improvement of 36.87% in training time using the distributed WTO method compared to the baseline with a single learning agent, and the accuracy increases 4.69% more using the proposed method in contrast with the baseline. As this is a preliminary study towards the distributed WTO method for training CNN models, we anticipate this approach can be used in robotics, multi-agent systems, federated learning, complex optimization problems, and many others, where an optimization task is required to be solved fast and accurate. © 2023 IEEE.
    Scopus© Citations 2  9  1