Examinando por Autor "Masegosa Arredondo, Antonio David"

Mostrando 1 - 5 de 5
  • Miniatura
    Ítem
    Emerging trends in machine learning assisted optimization techniques across intelligent transportation systems
    (Institute of Electrical and Electronics Engineers Inc., 2024) Itoro Afolayan, Blessing; Ghosh, Arka; Fajardo Calderín, Jenny; Masegosa Arredondo, Antonio David
    Artificial intelligence (AI) plays a critical role in Intelligent Transport Systems (ITS) as urban areas grow by processing data for safety enhancements, predictive analysis, and traffic management. This results in better traffic control, lower emissions, and preventative actions to lessen the effects of accidents. Despite these developments, there isn't a thorough academic analysis that covers a variety of optimization strategies for transportation AI models. By presenting an in-depth analysis of AI optimization methods and their uses in ITSs, this work seeks to close this knowledge gap and give academics important new information on possible directions for future research. Model-based optimization approaches, reinforcement learning techniques, model-predictive control techniques, and generative AI techniques are the four areas into which this study divides AI optimization techniques for the sake of structure, clarity, and comparative analysis. Subcategories of optimization techniques and their corresponding applications are explored, and each category is thoroughly addressed. Researchers will be better able to comprehend the state of AI optimization for transportation management today and in the future thanks to this methodical methodology. The most cutting-edge optimization methods created in the last five years are summarized in this review. This work acts as a compass for future research initiatives targeted at developing scalable and adaptable AI solutions for transportation management by identifying common approaches and highlighting research needs.
  • No hay miniatura disponible
    Ítem
    A graph CNN-LSTM neural network for short and long-term traffic forecasting based on trajectory data
    (Elsevier Ltd, 2020-03) Bogaerts, Toon; Masegosa Arredondo, Antonio David; Angarita Zapata, Juan S.; Onieva Caracuel, Enrique; Hellinckx, Peter
    Traffic forecasting is an important research area in Intelligent Transportation Systems that is focused on anticipating traffic in order to mitigate congestion. In this work we propose a deep neural network that simultaneously extracts the spatial features of traffic, using graph convolution, and its temporal features by means of Long Short Term Memory (LSTM) cells to make both short-term and long-term predictions. The model is trained and tested using sparse trajectory (GPS) data coming from the ride-hailing service of DiDi in the cities of Xi'an and Chengdu in China. Besides, presenting the deep neural network, we also propose a data-reduction technique based on temporal correlation to select the most relevant road links to be used as input. Combining the suggested approaches, our model obtains better results compared to high-performance algorithms for traffic forecasting, such as LSTM or the algorithms presented in the TRANSFOR19 forecasting competition. The model is capable of maintaining its performance over different time-horizons from 5 min to up to 4 h with multi-step predictions.
  • Miniatura
    Ítem
    A new Hyper-heuristic based on Adaptive Simulated Annealing and Reinforcement Learning for the Capacitated Electric Vehicle Routing Problem
    (Elsevier Ltd, 2024-10-15) Rodríguez Esparza, Erick; Masegosa Arredondo, Antonio David; Oliva, Diego; Onieva Caracuel, Enrique
    Electric vehicles (EVs) have been adopted in urban areas to reduce environmental pollution and global warming due to the increasing number of freight vehicles. However, there are still deficiencies in routing the trajectories of last-mile logistics that continue to impact social and economic sustainability. For that reason, in this paper, a hyper-heuristic (HH) approach called Hyper-heuristic Adaptive Simulated Annealing with Reinforcement Learning (HHASARL) is proposed. It is composed of a multi-armed bandit method and the self-adaptive Simulated Annealing (SA) metaheuristic algorithm for solving the problem called Capacitated Electric Vehicle Routing Problem (CEVRP). Due to the limited number of charging stations and the travel range of EVs, the EVs must require battery recharging moments in advance and reduce travel times and costs. The implementation of the HH improves multiple minimum best-known solutions and obtains the best mean values for some high-dimensional instances for the proposed benchmark for the IEEE WCCI2020 competition.
  • Miniatura
    Ítem
    Optimizing road traffic surveillance: a robust hyper-heuristic approach for vehicle segmentation
    (Institute of Electrical and Electronics Engineers Inc., 2024) Rodríguez Esparza, Erick; Ramos Soto, Oscar; Masegosa Arredondo, Antonio David; Onieva Caracuel, Enrique; Oliva, Diego; Arriandiaga Laresgoiti, Ander; Ghosh, Arka
    Due to rising consumer demand and traffic congestion, last-mile logistics is becoming more challenging. To optimize urban distribution networks, digital image processing plays a key role in addressing these challenges through efficient traffic monitoring systems, an essential component of intelligent transportation systems. This paper introduces the Hyper-heuristic Genetic Algorithm based on Thompson Sampling with Diversity (HHGATSD), a novel approach to efficiently solving complex optimization and versatility problems in image segmentation. We evaluate its efficiency and robustness using the IEEE CEC2017 benchmark function set in general optimization problems with 30 and 50 dimensions. HHGATSD's applicability extends beyond optimization to computer vision in traffic management. First, the multilevel thresholding segmentation is performed on images extracted from the Berkeley Segmentation Dataset with minimum cross-entropy as the objective function, and its performance is compared using PSNR, SSIM, and FSIM metrics. Following that, the proposed methodology addresses the task of vehicle segmentation in traffic camera videos, reaffirming HHGATSD's effectiveness, adaptability, and consistency by consistently outperforming alternative segmentation methods found in the state-of-the-art. The results of comprehensive experiments, validated by statistical and non-parametric analyses, show that the proposed hyper-heuristic and methodology produce accurate and consistent segmentations for road traffic surveillance compared to the other methods in the literature
  • Miniatura
    Ítem
    Understanding the role of diversity in ensemble-based AutoML methods for classification tasks
    (Institute of Electrical and Electronics Engineers Inc., 2025-04-17) Osei, Salomey; Masegosa, Andrés R.; Masegosa Arredondo, Antonio David
    Ensemble-based Automated Machine Learning (AutoML) methods have gained prominence for their ability to combine diverse machine learning models, achieving superior generalization performance. Despite their empirical success, the underlying mechanisms driving this performance, particularly the role of model diversity, are not yet adequately understood. This study uses novel theoretical frameworks related to the role of diversity in ensembles, which were recently proposed, to shed light on this issue. In this work, we focus on AutoML methods for classification tasks. We use AUTO-SKLEARN (a widely used AutoML ensemble-based method) as a basis. More specifically, we examine how individual model diversity and performance evolves across the four key phases of AUTO-SKLEARN (base-learners, meta-learning, Bayesian Optimization (BO), and Caruana Ensemble). We also examine how they contribute to the diversity and performance of the final ensemble produced by the AutoML method. Using datasets from the AutoML benchmark, we empirically validate these insights by analyzing error rates and diversity measures across the mentioned phases. Our findings highlight the trade-off between individual model accuracy and ensemble diversity, showing that phases like BO improve the mean error rate of classifiers by nearly 50% percent but reduce their mean diversity by 20%. However, the Caruana phase increases the diversity by a 50% compared to the BO phase, allowing better generalization despite the higher mean error rate of the selected individual models (48% higher than BO). This work provides theoretical and empirical evidence that diversity is critical to the success of ensemble-based AutoML methods and a deeper understanding of diversity’s impact on generalization performance and the role of the different AutoML phases. These findings can contribute to advance the development of more robust and theoretically grounded AutoML frameworks