Slide
Centre Interdisciplinaire
de Recherche et d’Innovation
en Cybersécurité et Société
de Recherche et d’Innovation
en Cybersécurité et Société
1.
Gapski, M. C. B.; Kawai, V. A. S.; Leticio, G. R.; Valem, L. P.; Pedronette, D. C. G.; Allili, M. S.
Graph Neural Networks for Semi-Supervised Image Classification with Multi-Feature Aggregation Article de journal
Dans: Journal of the Brazilian Computer Society, vol. 32, no 1, p. 1731–1754, 2026, ISSN: 01046500 (ISSN).
Résumé | Liens | BibTeX | Étiquettes: Classification accuracy, Convolution, Convolutional neural networks, extraction, Feature extraction, Feature extractor, Feature Fusion, Feature representation, Features fusions, Graph Neural Networks, Graph representation, Graph structures, Graph theory, Graphic methods, Image classification, Labeled data, Rank Aggregation, Semi-supervised, Semi-supervised image classification, Supervised image classifications, Textures
@article{gapskiGraphNeuralNetworks2026,
title = {Graph Neural Networks for Semi-Supervised Image Classification with Multi-Feature Aggregation},
author = {M. C. B. Gapski and V. A. S. Kawai and G. R. Leticio and L. P. Valem and D. C. G. Pedronette and M. S. Allili},
url = {https://www.scopus.com/pages/publications/105045179876?origin=resultslist},
doi = {10.5753/jbcs.2026.5880},
issn = {01046500 (ISSN)},
year = {2026},
date = {2026-01-01},
journal = {Journal of the Brazilian Computer Society},
volume = {32},
number = {1},
pages = {1731–1754},
publisher = {Brazilian Computing Society},
abstract = {Feature extraction involves the identification and extraction of salient characteristics or patterns, including edges, textures, shapes, and color attributes. Contemporary feature extractors predominantly leverage deep learning architectures, such as Convolutional Neural Networks (CNNs) and Vision Transformers (VITs). The availability of diverse feature extractors in the literature provides a wide range of feature representations. Features extracted from an image depend on the specific application, the chosen extractor, and its configuration. Therefore, integrating complementary information by combining distinct extractors offers a promising way to enhance performance. Graph Neural Networks (GNNs), particularly Graph Convolutional Networks (GCNs), have emerged as powerful and widely adopted approaches for semi-supervised image classification, as they effectively leverage both labeled and unlabeled data while exploiting the underlying graph structures that capture relationships among samples. This study proposes a novel approach for GNNs in scenarios where labeled data is scarce, by integrating diverse sets of feature and graph representations derived from various extractors in classification scenarios. Experimental investigations were conducted, encompassing combinations of distinct feature and graph extractors, as well as rank aggregation strategies. The primary contributions of this work are underscored by the experimental findings, which demonstrate that the strategic combination of feature and graph representations, coupled with the application of manifold learning for graph processing, leads to significant improvements in classification accuracy across the majority of experimental conditions. Furthermore, the utilization of rank aggregation techniques to integrate features from different extractors was shown to enhance classification accuracy. © 2026, Brazilian Computing Society. All rights reserved.},
keywords = {Classification accuracy, Convolution, Convolutional neural networks, extraction, Feature extraction, Feature extractor, Feature Fusion, Feature representation, Features fusions, Graph Neural Networks, Graph representation, Graph structures, Graph theory, Graphic methods, Image classification, Labeled data, Rank Aggregation, Semi-supervised, Semi-supervised image classification, Supervised image classifications, Textures},
pubstate = {published},
tppubtype = {article}
}
Feature extraction involves the identification and extraction of salient characteristics or patterns, including edges, textures, shapes, and color attributes. Contemporary feature extractors predominantly leverage deep learning architectures, such as Convolutional Neural Networks (CNNs) and Vision Transformers (VITs). The availability of diverse feature extractors in the literature provides a wide range of feature representations. Features extracted from an image depend on the specific application, the chosen extractor, and its configuration. Therefore, integrating complementary information by combining distinct extractors offers a promising way to enhance performance. Graph Neural Networks (GNNs), particularly Graph Convolutional Networks (GCNs), have emerged as powerful and widely adopted approaches for semi-supervised image classification, as they effectively leverage both labeled and unlabeled data while exploiting the underlying graph structures that capture relationships among samples. This study proposes a novel approach for GNNs in scenarios where labeled data is scarce, by integrating diverse sets of feature and graph representations derived from various extractors in classification scenarios. Experimental investigations were conducted, encompassing combinations of distinct feature and graph extractors, as well as rank aggregation strategies. The primary contributions of this work are underscored by the experimental findings, which demonstrate that the strategic combination of feature and graph representations, coupled with the application of manifold learning for graph processing, leads to significant improvements in classification accuracy across the majority of experimental conditions. Furthermore, the utilization of rank aggregation techniques to integrate features from different extractors was shown to enhance classification accuracy. © 2026, Brazilian Computing Society. All rights reserved.
2.
Nouboukpo, A.; Allili, M. S.
Semi-supervised flow-augmented Gaussian mixture VAEs with weighted contrastive learning for out-of-distribution detection Article de journal
Dans: Knowledge-Based Systems, vol. 347, 2026, ISSN: 09507051 (ISSN).
Résumé | Liens | BibTeX | Étiquettes: Auto encoders, Benchmarking, Classification (of information), Clusterings, Contrastive Learning, Flow based, Flow-based GMM clustering, Gaussian distribution, Gaussian-mixtures, Learning algorithms, Learning systems, Out-of-distribution detection, Semantics, Semi-supervised, Semi-supervised learning, uncertainty, Variational autoencoder, Variational autoencoders (VAEs)
@article{nouboukpoSemisupervisedFlowaugmentedGaussian2026,
title = {Semi-supervised flow-augmented Gaussian mixture VAEs with weighted contrastive learning for out-of-distribution detection},
author = {A. Nouboukpo and M. S. Allili},
url = {https://www.scopus.com/pages/publications/105040505017?origin=resultslist},
doi = {10.1016/j.knosys.2026.116312},
issn = {09507051 (ISSN)},
year = {2026},
date = {2026-01-01},
journal = {Knowledge-Based Systems},
volume = {347},
publisher = {Elsevier B.V.},
abstract = {Out-of-distribution (OOD) detection is essential for the safe deployment of machine learning systems, particularly in high-stakes domains where identifying inputs that deviate from the in-distribution (ID) is critical. However, in real-world scenarios, true OOD data are typically unavailable at training time, and only auxiliary datasets acting as imperfect proxies can be used. This makes fully supervised approaches impractical and potentially biased toward specific anomaly types. To address this limitation, we propose FLoW-ssGMVAE, a semi-supervised generative–discriminative framework that leverages limited proxy anomaly supervision together with abundant unlabeled data for robust OOD detection. Our model integrates a flow-augmented Gaussian Mixture VAE, enabling flexible latent modeling that captures nonlinear and multimodal intra-class distributions. Unlike standard VAEs with unimodal Gaussian priors, it uses class-conditional normalizing flows to better represent complex ID and OOD variability. To further refine the latent space and strengthen ID/OOD separation, our model integrates a weighted contrastive learning objective guided by sample-wise attention scores derived from likelihood uncertainty. This mechanism emphasizes ambiguous or hard-to-classify instances, reinforcing semantic boundaries and reducing false detections. By combining expressive generative modeling with uncertainty-aware discriminative training, our method constructs a structured and interpretable latent space that reliably identifies OOD samples under limited supervision. Experiments on standard benchmarks validate the effectiveness and scalability of our approach, demonstrating state-of-the-art performance with reasonable computational cost. Our implementation will be available at: FLoW-ssGMVAE. © 2026},
keywords = {Auto encoders, Benchmarking, Classification (of information), Clusterings, Contrastive Learning, Flow based, Flow-based GMM clustering, Gaussian distribution, Gaussian-mixtures, Learning algorithms, Learning systems, Out-of-distribution detection, Semantics, Semi-supervised, Semi-supervised learning, uncertainty, Variational autoencoder, Variational autoencoders (VAEs)},
pubstate = {published},
tppubtype = {article}
}
Out-of-distribution (OOD) detection is essential for the safe deployment of machine learning systems, particularly in high-stakes domains where identifying inputs that deviate from the in-distribution (ID) is critical. However, in real-world scenarios, true OOD data are typically unavailable at training time, and only auxiliary datasets acting as imperfect proxies can be used. This makes fully supervised approaches impractical and potentially biased toward specific anomaly types. To address this limitation, we propose FLoW-ssGMVAE, a semi-supervised generative–discriminative framework that leverages limited proxy anomaly supervision together with abundant unlabeled data for robust OOD detection. Our model integrates a flow-augmented Gaussian Mixture VAE, enabling flexible latent modeling that captures nonlinear and multimodal intra-class distributions. Unlike standard VAEs with unimodal Gaussian priors, it uses class-conditional normalizing flows to better represent complex ID and OOD variability. To further refine the latent space and strengthen ID/OOD separation, our model integrates a weighted contrastive learning objective guided by sample-wise attention scores derived from likelihood uncertainty. This mechanism emphasizes ambiguous or hard-to-classify instances, reinforcing semantic boundaries and reducing false detections. By combining expressive generative modeling with uncertainty-aware discriminative training, our method constructs a structured and interpretable latent space that reliably identifies OOD samples under limited supervision. Experiments on standard benchmarks validate the effectiveness and scalability of our approach, demonstrating state-of-the-art performance with reasonable computational cost. Our implementation will be available at: FLoW-ssGMVAE. © 2026
3.
Nouboukpo, A.; Allaoui, M. L.; Allili, M. S.
Multi-scale spatial consistency for deep semi-supervised skin lesion segmentation Article de journal
Dans: Engineering Applications of Artificial Intelligence, vol. 135, 2024, ISSN: 09521976 (ISSN), (Publisher: Elsevier Ltd).
Résumé | Liens | BibTeX | Étiquettes: Deep learning, Dermatology, Image segmentation, Lesion segmentations, Medical imaging, Multi-scales, Semi-supervised, Semi-supervised learning, Skin lesion, Skin lesion segmentation, Spatial consistency, Spatially constrained mixture model, Spatially-constrained mixture models, Supervised learning, UNets, Unlabeled data
@article{nouboukpo_multi-scale_2024,
title = {Multi-scale spatial consistency for deep semi-supervised skin lesion segmentation},
author = {A. Nouboukpo and M. L. Allaoui and M. S. Allili},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85195700182&doi=10.1016%2fj.engappai.2024.108681&partnerID=40&md5=e1cc2b6a1bb0aed530e8c04583c76167},
doi = {10.1016/j.engappai.2024.108681},
issn = {09521976 (ISSN)},
year = {2024},
date = {2024-01-01},
journal = {Engineering Applications of Artificial Intelligence},
volume = {135},
publisher = {Elsevier Ltd},
abstract = {This paper introduces a novel semi-supervised framework, the Multiscale Spatial Consistency Network (MSCNet), for robust semi-supervised skin lesion segmentation. MSCNet uses local and global spatial consistency to leverage a minimal set of labeled data, supplemented by a large number of unlabeled data, to improve segmentation. The model is is based on a single Encoder–Decoder (ED) network, augmented with a Spatially-Constrained Mixture Model (SCMM) to enforce spatial coherence in predictions. To encode the local spatial consistency, a hierarchical superpixel structure is used capture local region context (LRC), bolstering the model capacity to discern fine-grained lesion details. Global consistency is enforced through the SCMM module, which uses a larger context for lesion/background discrimination. In addition, it enables efficient leveraging of the unlabeled data through pseudo-label generation. Experiments demonstrate that the MSCNet outperforms existing state-of-the-art methods in segmenting complex lesions. The MSCNet has an excellent generalization capability, offering a promising direction for semi-supervised medical image segmentation, particularly in scenarios with limited annotated data. The code is available at https://github.com/AdamaTG/MSCNet. © 2024 Elsevier Ltd},
note = {Publisher: Elsevier Ltd},
keywords = {Deep learning, Dermatology, Image segmentation, Lesion segmentations, Medical imaging, Multi-scales, Semi-supervised, Semi-supervised learning, Skin lesion, Skin lesion segmentation, Spatial consistency, Spatially constrained mixture model, Spatially-constrained mixture models, Supervised learning, UNets, Unlabeled data},
pubstate = {published},
tppubtype = {article}
}
This paper introduces a novel semi-supervised framework, the Multiscale Spatial Consistency Network (MSCNet), for robust semi-supervised skin lesion segmentation. MSCNet uses local and global spatial consistency to leverage a minimal set of labeled data, supplemented by a large number of unlabeled data, to improve segmentation. The model is is based on a single Encoder–Decoder (ED) network, augmented with a Spatially-Constrained Mixture Model (SCMM) to enforce spatial coherence in predictions. To encode the local spatial consistency, a hierarchical superpixel structure is used capture local region context (LRC), bolstering the model capacity to discern fine-grained lesion details. Global consistency is enforced through the SCMM module, which uses a larger context for lesion/background discrimination. In addition, it enables efficient leveraging of the unlabeled data through pseudo-label generation. Experiments demonstrate that the MSCNet outperforms existing state-of-the-art methods in segmenting complex lesions. The MSCNet has an excellent generalization capability, offering a promising direction for semi-supervised medical image segmentation, particularly in scenarios with limited annotated data. The code is available at https://github.com/AdamaTG/MSCNet. © 2024 Elsevier Ltd
4.
Nouboukpo, A.; Allaoui, M. L.; Allili, M. S.
Multi-scale spatial consistency for deep semi-supervised skin lesion segmentation Article de journal
Dans: Engineering Applications of Artificial Intelligence, vol. 135, 2024, ISSN: 09521976 (ISSN).
Résumé | Liens | BibTeX | Étiquettes: Deep learning, Dermatology, Image segmentation, Lesion segmentations, Medical imaging, Multi-scales, Semi-supervised, Semi-supervised learning, Skin lesion, Skin lesion segmentation, Spatial consistency, Spatially constrained mixture model, Spatially-constrained mixture models, Supervised learning, UNets, Unlabeled data
@article{nouboukpoMultiscaleSpatialConsistency2024,
title = {Multi-scale spatial consistency for deep semi-supervised skin lesion segmentation},
author = {A. Nouboukpo and M. L. Allaoui and M. S. Allili},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85195700182&doi=10.1016%2fj.engappai.2024.108681&partnerID=40&md5=e1cc2b6a1bb0aed530e8c04583c76167},
doi = {10.1016/j.engappai.2024.108681},
issn = {09521976 (ISSN)},
year = {2024},
date = {2024-01-01},
journal = {Engineering Applications of Artificial Intelligence},
volume = {135},
publisher = {Elsevier Ltd},
abstract = {This paper introduces a novel semi-supervised framework, the Multiscale Spatial Consistency Network (MSCNet), for robust semi-supervised skin lesion segmentation. MSCNet uses local and global spatial consistency to leverage a minimal set of labeled data, supplemented by a large number of unlabeled data, to improve segmentation. The model is is based on a single Encoder–Decoder (ED) network, augmented with a Spatially-Constrained Mixture Model (SCMM) to enforce spatial coherence in predictions. To encode the local spatial consistency, a hierarchical superpixel structure is used capture local region context (LRC), bolstering the model capacity to discern fine-grained lesion details. Global consistency is enforced through the SCMM module, which uses a larger context for lesion/background discrimination. In addition, it enables efficient leveraging of the unlabeled data through pseudo-label generation. Experiments demonstrate that the MSCNet outperforms existing state-of-the-art methods in segmenting complex lesions. The MSCNet has an excellent generalization capability, offering a promising direction for semi-supervised medical image segmentation, particularly in scenarios with limited annotated data. The code is available at https://github.com/AdamaTG/MSCNet. © 2024 Elsevier Ltd},
keywords = {Deep learning, Dermatology, Image segmentation, Lesion segmentations, Medical imaging, Multi-scales, Semi-supervised, Semi-supervised learning, Skin lesion, Skin lesion segmentation, Spatial consistency, Spatially constrained mixture model, Spatially-constrained mixture models, Supervised learning, UNets, Unlabeled data},
pubstate = {published},
tppubtype = {article}
}
This paper introduces a novel semi-supervised framework, the Multiscale Spatial Consistency Network (MSCNet), for robust semi-supervised skin lesion segmentation. MSCNet uses local and global spatial consistency to leverage a minimal set of labeled data, supplemented by a large number of unlabeled data, to improve segmentation. The model is is based on a single Encoder–Decoder (ED) network, augmented with a Spatially-Constrained Mixture Model (SCMM) to enforce spatial coherence in predictions. To encode the local spatial consistency, a hierarchical superpixel structure is used capture local region context (LRC), bolstering the model capacity to discern fine-grained lesion details. Global consistency is enforced through the SCMM module, which uses a larger context for lesion/background discrimination. In addition, it enables efficient leveraging of the unlabeled data through pseudo-label generation. Experiments demonstrate that the MSCNet outperforms existing state-of-the-art methods in segmenting complex lesions. The MSCNet has an excellent generalization capability, offering a promising direction for semi-supervised medical image segmentation, particularly in scenarios with limited annotated data. The code is available at https://github.com/AdamaTG/MSCNet. © 2024 Elsevier Ltd