Azedine Ouhmid
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Azedine Ouhmid has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Karim El Moutaouakil
Laboratory of Mathematics and Data Science, Polydisciplinary of Taza, Sidi Mohamed Ben Abdellah University, Taza 30000, Morocco
3 shared publications - Alina-Mihaela Patriciu
Department of Mathematics and Computer Science, Faculty of Sciences and Environment, Dunărea de Jos University of Galaţi, 800201 Galaţi, Romania
1 shared publication - Charchaoui Wiam
Laboratory of Mathematics and Data Science, Faculty of Applied Sciences of Taza, Sidi Mohamed Ben Abdellah University, Taza 35000, Morocco
1 shared publication - Fatima Belhabib
Engineering Science Laboratory, Mohamed Ben Abdellah University,ISPITS , FES, Morocco
1 shared publication - Mohamed Barhdadi
Research Center, High Studies of Engineering School, Oujda 60000, Morocco
1 shared publication - Mohammed Barrouch
Laboratory of Mathematics and Data Science, Polydisciplinary Faculty of Taza, Sidi Mohamed Ben Abdellah University, Taza 35000, Morocco
1 shared publication
Latest Publications
- Open Access
- 2 November 2026
- Published by MDPI AG in Information
- Vol. 17 (11) , 1092
Omni-Flex GraphFormer: Parallel Fusion of Local Graph Aggregation and Global Attention Across Image Resolutions
Graph neural networks (GNNs) capture local image structure with few parameters, whereas Vision Transformers (ViTs) model global context at a higher computational cost. We propose the Omni-Flex GraphFormer, which applies a graph branch (OMNIVIG: max-relative aggregation over a fixed pixel neighbourhood) and a Transformer branch (FLEXIVIT: global multi-head self-attention) in parallel and sums their outputs. Because each baseline is the hybrid with one branch removed, comparing the three models on eight datasets from to pixels, under equal training budgets and over three seeds, is a controlled branch ablation. Global attention is decisive: the hybrid exceeds OMNIVIG in test macro-F1 on every dataset and in all 24 runs, by to points (Wilcoxon test, ). Against FLEXIVIT, the hybrid has the best mean F1 on six datasets and a 40– smaller seed-to-seed spread on CIFAR-10 and CIFAR-100, but the difference is not significant (). The graph branch adds only – to the number of multiply–accumulate operations, and under warmed-up timing, both models run at the same speed, whereas unwarmed timing overstates latency by up to . Under Gaussian noise, the summed hybrid degrades like the graph branch, which motivates gated or attention-guiding fusion.- 2 November 2026
- Published by MDPI AG in Information
- Vol. 17 (11) , 1092
Open AccessOmni-Flex GraphFormer: Parallel Fusion of Local Graph Aggregation and Global Attention Across Image Resolutions
Graph neural networks (GNNs) capture local image structure with few parameters, whereas Vision Transformers (ViTs) model global context at a higher computational cost. We propose the Omni-Flex GraphFormer, which applies a graph branch (OMNIVIG: max-relative aggregation over a fixed pixel neighbourhood) and a Transformer branch (FLEXIVIT: global multi-head self-attention) in parallel and sums their outputs. Because each baseline is the hybrid with one branch removed, comparing the three models on eight datasets from to pixels, under equal training budgets and over three seeds, is a controlled branch ablation. Global attention is decisive: the hybrid exceeds OMNIVIG in test macro-F1 on every dataset and in all 24 runs, by to points (Wilcoxon test, ). Against FLEXIVIT, the hybrid has the best mean F1 on six datasets and a 40– smaller seed-to-seed spread on CIFAR-10 and CIFAR-100, but the difference is not significant (). The graph branch adds only – to the number of multiply–accumulate operations, and under warmed-up timing, both models run at the same speed, whereas unwarmed timing overstates latency by up to . Under Gaussian noise, the summed hybrid degrades like the graph branch, which motivates gated or attention-guiding fusion. - Open Access
Decomposition and Symmetric Kernel Deep Neural Network Fuzzy Support Vector Machine
Algorithms involving kernel functions, such as support vector machine (SVM), have attracted huge attention within the artificial learning communities. The performance of these algorithms is greatly influenced by outliers and the choice of kernel functions. This paper introduces a new version of SVM named Deep Decomposition Neural Network Fuzzy SVM (DDNN-FSVM). To this end, we consider an auto-encoder (AE) deep neural network with three layers: input, hidden, and output. Unusually, the AE’s hidden layer comprises a number of neurons greater than the dimension of the input samples, which guarantees linear data separation. The encoder operator is then introduced into the FSVM’s dual to map the training samples to high-dimension spaces. To learn the support vectors and autoencoder parameters, we introduce the loss function and regularization terms in the FSVM dual. To learn from large-scale data, we decompose the resulting model into three small-dimensional submodels using Lagrangian decomposition. To solve the resulting problems, we use SMO, ISDA, and SCG for optimization problems involving large-scale data. We demonstrate that the optimal values of the three submodels solved in parallel provide a good lower bound for the optimal value of the initial model. In addition, thanks to its use of fuzzy weights, DDNN-FSVM is resistant to outliers. Moreover, DDNN-FSVM simultaneously learns the appropriate kernel function and separation path. We tested DDNN-FSVM on several well-known digital and image datasets and compared it to well-known classifiers on the basis of accuracy, precision, f-measure, g-means, and recall. On average, DDNN-FSVM improved on the performance of the classic FSVM across all datasets and outperformed several well-known classifiers.
Open AccessDecomposition and Symmetric Kernel Deep Neural Network Fuzzy Support Vector Machine
Algorithms involving kernel functions, such as support vector machine (SVM), have attracted huge attention within the artificial learning communities. The performance of these algorithms is greatly influenced by outliers and the choice of kernel functions. This paper introduces a new version of SVM named Deep Decomposition Neural Network Fuzzy SVM (DDNN-FSVM). To this end, we consider an auto-encoder (AE) deep neural network with three layers: input, hidden, and output. Unusually, the AE’s hidden layer comprises a number of neurons greater than the dimension of the input samples, which guarantees linear data separation. The encoder operator is then introduced into the FSVM’s dual to map the training samples to high-dimension spaces. To learn the support vectors and autoencoder parameters, we introduce the loss function and regularization terms in the FSVM dual. To learn from large-scale data, we decompose the resulting model into three small-dimensional submodels using Lagrangian decomposition. To solve the resulting problems, we use SMO, ISDA, and SCG for optimization problems involving large-scale data. We demonstrate that the optimal values of the three submodels solved in parallel provide a good lower bound for the optimal value of the initial model. In addition, thanks to its use of fuzzy weights, DDNN-FSVM is resistant to outliers. Moreover, DDNN-FSVM simultaneously learns the appropriate kernel function and separation path. We tested DDNN-FSVM on several well-known digital and image datasets and compared it to well-known classifiers on the basis of accuracy, precision, f-measure, g-means, and recall. On average, DDNN-FSVM improved on the performance of the classic FSVM across all datasets and outperformed several well-known classifiers.
- Open Access
Enhancing Symmetry and Memory in the Fractional Economic Growing Quantity (FEGQ) Model
In this paper, we present a novel approach to inventory management modeling, specifically tailored for growing items. We extend traditional economic growth quantity (EGQ) models by introducing the fractional economic growing quantity (FEGQ) model. This new approach improves the model’s symmetry and dynamic responsiveness, providing a more precise representation of the changing nature of inventory items. Additionally, the use of fractional derivatives allows our model to incorporate the memory effect, introducing a new dynamic concept in inventory management. This advancement enables us to select the optimal business policy to maximize profit. We adopt the fractional derivative in terms of Caputo derivative sense to model the inventory level associated with the items. To analytically solve the (FEGQ) model, we use the Laplacian transform to obtain an algebraic equation. As for the logistic function, known for its symmetrical S-shaped curve, it closely mirrors real-life growth patterns and is defined using fractional calculus. We apply an iterative approximation method, specifically the Adomian decomposition method, to solve the fractional logistic function. Through a sensitivity analysis, we delve for the first time into the discussion of the initial weights, which have a massive impact on the total profit level. The provided numerical data indicate that the firm began with a favorable policy. In the following years, several misguided practices were implemented that led to a decrease in profitability. The healing process began once again by selecting more effective strategies.
Open AccessEnhancing Symmetry and Memory in the Fractional Economic Growing Quantity (FEGQ) Model
In this paper, we present a novel approach to inventory management modeling, specifically tailored for growing items. We extend traditional economic growth quantity (EGQ) models by introducing the fractional economic growing quantity (FEGQ) model. This new approach improves the model’s symmetry and dynamic responsiveness, providing a more precise representation of the changing nature of inventory items. Additionally, the use of fractional derivatives allows our model to incorporate the memory effect, introducing a new dynamic concept in inventory management. This advancement enables us to select the optimal business policy to maximize profit. We adopt the fractional derivative in terms of Caputo derivative sense to model the inventory level associated with the items. To analytically solve the (FEGQ) model, we use the Laplacian transform to obtain an algebraic equation. As for the logistic function, known for its symmetrical S-shaped curve, it closely mirrors real-life growth patterns and is defined using fractional calculus. We apply an iterative approximation method, specifically the Adomian decomposition method, to solve the fractional logistic function. Through a sensitivity analysis, we delve for the first time into the discussion of the initial weights, which have a massive impact on the total profit level. The provided numerical data indicate that the firm began with a favorable policy. In the following years, several misguided practices were implemented that led to a decrease in profitability. The healing process began once again by selecting more effective strategies.