Driss, Maryam Ben; Sabir, Essaid; Elbiaze, Halima; Diallo, Abloulaye Baniré
Fast & Energy Efficient Federated Learning Using Multi-Attribute Client Clustering and Selection [External] Article d'actes
Dans: Oslo, Norway, 2025.
Résumé | Liens | BibTeX | Étiquettes: Article de conference
@inproceedings{ben_driss_fast_2025,
title = {Fast & Energy Efficient Federated Learning Using Multi-Attribute Client Clustering and Selection},
author = {Maryam Ben Driss and Essaid Sabir and Halima Elbiaze and Abloulaye Baniré Diallo},
url = {https://ieeexplore.ieee.org/document/11174476},
year = {2025},
date = {2025-06-01},
address = {Oslo, Norway},
abstract = {Federated Learning (FL) presents a promising paradigm for decentralized model training; however, its real-world adoption is hindered by several critical challenges, including non-independent and identically distributed (non-IID) data across clients, heterogeneous computational capabilities, and significant communication overhead. To address these issues, this paper introduces a novel multi-attribute client clustering and selection framework for FL. The proposed approach groups clients according to data distribution, device capabilities, geographic location, and model update behavior. Within each cluster, an adaptive client selection mechanism leverages dynamic attributes such as residual energy, data freshness, and client participation motivation to identify the most suitable participants. Experimental evaluations on standard FL benchmark datasets demonstrate that the proposed framework achieves faster convergence, higher global model accuracy, and improved energy efficiency compared to state-of-the-art approaches.},
keywords = {Article de conference},
pubstate = {published},
tppubtype = {inproceedings}
}
Federated Learning (FL) presents a promising paradigm for decentralized model training; however, its real-world adoption is hindered by several critical challenges, including non-independent and identically distributed (non-IID) data across clients, heterogeneous computational capabilities, and significant communication overhead. To address these issues, this paper introduces a novel multi-attribute client clustering and selection framework for FL. The proposed approach groups clients according to data distribution, device capabilities, geographic location, and model update behavior. Within each cluster, an adaptive client selection mechanism leverages dynamic attributes such as residual energy, data freshness, and client participation motivation to identify the most suitable participants. Experimental evaluations on standard FL benchmark datasets demonstrate that the proposed framework achieves faster convergence, higher global model accuracy, and improved energy efficiency compared to state-of-the-art approaches.
Massoua, Armand Bandiang; Diallo, Abdoulaye Banire; Bouguessa, Mohamed
Towards Robust Time-to-Event Prediction: Integrating the Variational Information Bottleneck with Neural Survival Model [External] Article d'actes
Dans: 2024 International Joint Conference on Neural Networks (IJCNN), p. 1–8, IEEE, Yokohama, Japan, 2024, ISBN: 979-8-3503-5931-2.
Liens | BibTeX | Étiquettes: Article de conference
@inproceedings{massoua_towards_2024,
title = {Towards Robust Time-to-Event Prediction: Integrating the Variational Information Bottleneck with Neural Survival Model},
author = {Armand Bandiang Massoua and Abdoulaye Banire Diallo and Mohamed Bouguessa},
url = {https://ieeexplore.ieee.org/document/10651066/},
doi = {10.1109/IJCNN60899.2024.10651066},
isbn = {979-8-3503-5931-2},
year = {2024},
date = {2024-06-01},
urldate = {2024-10-16},
booktitle = {2024 International Joint Conference on Neural Networks (IJCNN)},
pages = {1–8},
publisher = {IEEE},
address = {Yokohama, Japan},
keywords = {Article de conference},
pubstate = {published},
tppubtype = {inproceedings}
}
driss, Maryam Ben; Sabir, Essaid; Elbiaze, Halima; banire Diallo, Abdoulaye; Sadik, Mohamed
GWO-Boosted Multi-Attribute Client Selection for Over-The-Air Federated Learning [External] Article d'actes
Dans: IEEE, Montreal, Canada, 2024.
Liens | BibTeX | Étiquettes: Article de conference
@inproceedings{ben_driss_gwo-boosted_2024,
title = {GWO-Boosted Multi-Attribute Client Selection for Over-The-Air Federated Learning},
author = {Maryam Ben driss and Essaid Sabir and Halima Elbiaze and Abdoulaye banire Diallo and Mohamed Sadik},
url = {https://r-libre.teluq.ca/3274/1/1570983885%20stamped-e.pdf},
year = {2024},
date = {2024-05-01},
publisher = {IEEE},
address = {Montreal, Canada},
keywords = {Article de conference},
pubstate = {published},
tppubtype = {inproceedings}
}
Mrabah, Nairouz; Amar, Mohamed Mahmoud; Bouguessa, Mohamed; Diallo, Abdoulaye Banire
Toward Convex Manifolds: A Geometric Perspective for Deep Graph Clustering of Single-cell RNA-seq Data [External] Article d'actes
Dans: Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence, p. 4855–4863, International Joint Conferences on Artificial Intelligence Organization, Macau, SAR China, 2023, ISBN: 978-1-956792-03-4.
Résumé | Liens | BibTeX | Étiquettes: Article de conference
@inproceedings{mrabah_toward_2023,
title = {Toward Convex Manifolds: A Geometric Perspective for Deep Graph Clustering of Single-cell RNA-seq Data},
author = {Nairouz Mrabah and Mohamed Mahmoud Amar and Mohamed Bouguessa and Abdoulaye Banire Diallo},
url = {https://www.ijcai.org/proceedings/2023/540},
doi = {10.24963/ijcai.2023/540},
isbn = {978-1-956792-03-4},
year = {2023},
date = {2023-08-01},
urldate = {2024-10-16},
booktitle = {Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence},
pages = {4855–4863},
publisher = {International Joint Conferences on Artificial Intelligence Organization},
address = {Macau, SAR China},
abstract = {The deep clustering paradigm has shown great potential for discovering complex patterns that can reveal cell heterogeneity in single-cell RNA sequencing data. This paradigm involves two training phases: pretraining based on a pretext task and fine-tuning using pseudo-labels. Although current models yield promising results, they overlook the geometric distortions that regularly occur during the training process. More precisely, the transition between the two phases results in a coarse flattening of the latent structures, which can deteriorate the clustering performance. In this context, existing methods perform euclidean-based embedding clustering without ensuring the flatness and convexity of the latent manifolds. To address this problem, we incorporate two mechanisms. First, we introduce an overclustering loss to flatten the local curves. Second, we propose an adversarial mechanism to adjust the global geometric configuration. The second mechanism gradually transforms the latent structures into convex ones. Empirical results on a variety of gene expression datasets show that our model outperforms state-of-the-art methods.},
keywords = {Article de conference},
pubstate = {published},
tppubtype = {inproceedings}
}
The deep clustering paradigm has shown great potential for discovering complex patterns that can reveal cell heterogeneity in single-cell RNA sequencing data. This paradigm involves two training phases: pretraining based on a pretext task and fine-tuning using pseudo-labels. Although current models yield promising results, they overlook the geometric distortions that regularly occur during the training process. More precisely, the transition between the two phases results in a coarse flattening of the latent structures, which can deteriorate the clustering performance. In this context, existing methods perform euclidean-based embedding clustering without ensuring the flatness and convexity of the latent manifolds. To address this problem, we incorporate two mechanisms. First, we introduce an overclustering loss to flatten the local curves. Second, we propose an adversarial mechanism to adjust the global geometric configuration. The second mechanism gradually transforms the latent structures into convex ones. Empirical results on a variety of gene expression datasets show that our model outperforms state-of-the-art methods.
Mrabah, Nairouz; Amar, Mohamed Mahmoud; Bouguessa, Mohamed; Diallo, Abdoulaye Banire
Exploring the Interaction between Local and Global Latent Configurations for Clustering Single-Cell RNA-Seq: A Unified Perspective [External] Article d'actes
Dans: Proceedings of the AAAI Conference on Artificial Intelligence, p. 9235–9242, 2023.
Résumé | Liens | BibTeX | Étiquettes: Article de conference
@inproceedings{mrabah_exploring_2023,
title = {Exploring the Interaction between Local and Global Latent Configurations for Clustering Single-Cell RNA-Seq: A Unified Perspective},
author = {Nairouz Mrabah and Mohamed Mahmoud Amar and Mohamed Bouguessa and Abdoulaye Banire Diallo},
url = {https://ojs.aaai.org/index.php/AAAI/article/view/26107},
doi = {10.1609/aaai.v37i8.26107},
year = {2023},
date = {2023-06-01},
urldate = {2024-10-16},
booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
volume = {37},
pages = {9235–9242},
abstract = {The most recent approaches for clustering single-cell RNA-sequencing data rely on deep auto-encoders. However, three major challenges remain unaddressed. First, current models overlook the impact of the cumulative errors induced by the pseudo-supervised embedding clustering task (Feature Randomness). Second, existing methods neglect the effect of the strong competition between embedding clustering and reconstruction (Feature Drift). Third, the previous deep clustering models regularly fail to consider the topological information of the latent data, even though the local and global latent configurations can bring complementary views to the clustering task. To address these challenges, we propose a novel approach that explores the interaction between local and global latent configurations to progressively adjust the reconstruction and embedding clustering tasks. We elaborate a topological and probabilistic filter to mitigate Feature Randomness and a cell-cell graph structure and content correction mechanism to counteract Feature Drift. The Zero-Inflated Negative Binomial model is also integrated to capture the characteristics of gene expression profiles. We conduct detailed experiments on real-world datasets from multiple representative genome sequencing platforms. Our approach outperforms the state-of-the-art clustering methods in various evaluation metrics.},
keywords = {Article de conference},
pubstate = {published},
tppubtype = {inproceedings}
}
The most recent approaches for clustering single-cell RNA-sequencing data rely on deep auto-encoders. However, three major challenges remain unaddressed. First, current models overlook the impact of the cumulative errors induced by the pseudo-supervised embedding clustering task (Feature Randomness). Second, existing methods neglect the effect of the strong competition between embedding clustering and reconstruction (Feature Drift). Third, the previous deep clustering models regularly fail to consider the topological information of the latent data, even though the local and global latent configurations can bring complementary views to the clustering task. To address these challenges, we propose a novel approach that explores the interaction between local and global latent configurations to progressively adjust the reconstruction and embedding clustering tasks. We elaborate a topological and probabilistic filter to mitigate Feature Randomness and a cell-cell graph structure and content correction mechanism to counteract Feature Drift. The Zero-Inflated Negative Binomial model is also integrated to capture the characteristics of gene expression profiles. We conduct detailed experiments on real-world datasets from multiple representative genome sequencing platforms. Our approach outperforms the state-of-the-art clustering methods in various evaluation metrics.
Naghashi, V.; Dallago, G. M.; Diallo, A. B.; Boukadoum, M.
Univariate and multivariate time-series methods to forecast dairy income [External] Article d'actes
Dans: Washington, DC, 2023.
Liens | BibTeX | Étiquettes: Article de conference
@inproceedings{naghashi_univariate_2023,
title = {Univariate and multivariate time-series methods to forecast dairy income},
author = {V. Naghashi and G. M. Dallago and A. B. Diallo and M. Boukadoum},
url = {https://openreview.net/pdf?id=0sGHJV7tRM},
year = {2023},
date = {2023-02-01},
address = {Washington, DC},
keywords = {Article de conference},
pubstate = {published},
tppubtype = {inproceedings}
}
Naghashi, Vahid; Diallo, Abdoulaye Banire
A Model for the Prediction of Lifetime Profit Estimate of Dairy Cattle (Student Abstract) [External] Article d'actes
Dans: Proceedings of the AAAI Conference on Artificial Intelligence, p. 13021–13022, 2022.
Résumé | Liens | BibTeX | Étiquettes: Article de conference
@inproceedings{naghashi_model_2022,
title = {A Model for the Prediction of Lifetime Profit Estimate of Dairy Cattle (Student Abstract)},
author = {Vahid Naghashi and Abdoulaye Banire Diallo},
url = {https://ojs.aaai.org/index.php/AAAI/article/view/21647},
doi = {10.1609/aaai.v36i11.21647},
year = {2022},
date = {2022-06-01},
urldate = {2024-10-16},
booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
volume = {36},
pages = {13021–13022},
abstract = {In livestock management, the decision of animal replacement requires an estimation of the lifetime profit of the animal based on multiple factors and operational conditions. In Dairy farms, this can be associated with the profit corresponding to milk production, health condition and herd management costs, which in turn may be a function of other factors including genetics and weather conditions. Estimating the profit of a cow can be expressed as a spatio-temporal problem where knowing the first batch of production (early-profit) can allow to predict the future batch of productions (late-profit).
This problem can be addressed either by a univariate or multivariate time series forecasting. Several approaches have been designed for time series forecasting including Auto-Regressive approaches, Recurrent Neural Network including Long Short Term Memory (LSTM) method and a very deep stack of fully-connected layers. In this paper, we proposed a LSTM based approach coupled with attention and linear layers to better capture the dairy features. We compare the model, with three other architectures including NBEATs, ARIMA, MUMU-RNN using dairy production of 292181 dairy cows. The results highlight the performence of the proposed model of the compared architectures. They also show that a univariate NBEATs could perform better than the multi-variate approach there are compared to. We also highlight that such architecture could allow to predict late-profit with an error less than 3$ per month, opening the way of better resource management in the dairy industry.},
keywords = {Article de conference},
pubstate = {published},
tppubtype = {inproceedings}
}
In livestock management, the decision of animal replacement requires an estimation of the lifetime profit of the animal based on multiple factors and operational conditions. In Dairy farms, this can be associated with the profit corresponding to milk production, health condition and herd management costs, which in turn may be a function of other factors including genetics and weather conditions. Estimating the profit of a cow can be expressed as a spatio-temporal problem where knowing the first batch of production (early-profit) can allow to predict the future batch of productions (late-profit).
This problem can be addressed either by a univariate or multivariate time series forecasting. Several approaches have been designed for time series forecasting including Auto-Regressive approaches, Recurrent Neural Network including Long Short Term Memory (LSTM) method and a very deep stack of fully-connected layers. In this paper, we proposed a LSTM based approach coupled with attention and linear layers to better capture the dairy features. We compare the model, with three other architectures including NBEATs, ARIMA, MUMU-RNN using dairy production of 292181 dairy cows. The results highlight the performence of the proposed model of the compared architectures. They also show that a univariate NBEATs could perform better than the multi-variate approach there are compared to. We also highlight that such architecture could allow to predict late-profit with an error less than 3$ per month, opening the way of better resource management in the dairy industry.
This problem can be addressed either by a univariate or multivariate time series forecasting. Several approaches have been designed for time series forecasting including Auto-Regressive approaches, Recurrent Neural Network including Long Short Term Memory (LSTM) method and a very deep stack of fully-connected layers. In this paper, we proposed a LSTM based approach coupled with attention and linear layers to better capture the dairy features. We compare the model, with three other architectures including NBEATs, ARIMA, MUMU-RNN using dairy production of 292181 dairy cows. The results highlight the performence of the proposed model of the compared architectures. They also show that a univariate NBEATs could perform better than the multi-variate approach there are compared to. We also highlight that such architecture could allow to predict late-profit with an error less than 3$ per month, opening the way of better resource management in the dairy industry.
Fuentes, V.; Martin, T.; Valtchev, P.; Diallo, A. B.; Lacroix, R.; Leduc, M.
DCPO: The dairy cattle performance ontology, a tool for domain modelling and data analytics [External] Article d'actes
Dans: Proceedings ICAR Annual Conference 2022 in Montreal ICAR Technical Series #26.pdf, Montreal, 2022.
Liens | BibTeX | Étiquettes: Article de conference
@inproceedings{fuentes_dcpo_2022,
title = {DCPO: The dairy cattle performance ontology, a tool for domain modelling and data analytics},
author = {V. Fuentes and T. Martin and P. Valtchev and A. B. Diallo and R. Lacroix and M. Leduc},
url = {https://www.icar.org/Documents/technical_series/ICAR-Technical-Series-no-26-Montreal/11%20DCPO%20Dairy%20cattle%20performance%20ontology.pdf},
year = {2022},
date = {2022-05-01},
booktitle = {Proceedings ICAR Annual Conference 2022 in Montreal ICAR Technical Series #26.pdf},
address = {Montreal},
keywords = {Article de conference},
pubstate = {published},
tppubtype = {inproceedings}
}
Ayat, M.; Bisson, G.; Prince, J.; Fuentes, V.; Warner, D.; Lefebvre, D. M.; Santschi, D. E.; Lacroix, R.
Automated anomaly detection for milk components and diagnostics in dairy herds [External] Article d'actes
Dans: Proceedings ICAR Annual Conference 2022 in Montreal ICAR Technical Series #26, Montreal, 2022.
Liens | BibTeX | Étiquettes: Article de conference
@inproceedings{ayat_automated_2022,
title = {Automated anomaly detection for milk components and diagnostics in dairy herds},
author = {M. Ayat and G. Bisson and J. Prince and V. Fuentes and D. Warner and D. M. Lefebvre and D. E. Santschi and R. Lacroix},
url = {https://www.icar.org/Documents/technical_series/ICAR-Technical-Series-no-26-Montreal/19%20Automated%20anomaly%20detection%20for%20milk%20components.pdf},
year = {2022},
date = {2022-05-01},
booktitle = {Proceedings ICAR Annual Conference 2022 in Montreal ICAR Technical Series #26},
address = {Montreal},
keywords = {Article de conference},
pubstate = {published},
tppubtype = {inproceedings}
}