Institution profile

University of Alabama at Birmingham

Academic institutionnorthamerica · us
Official website
Research library57linked papers
Opportunities0open roles
Selected work

Representative Papers

Predicting Anemia Among Under-Five Children in Nepal Using Machine Learning and Deep Learning

Feb 01, 2026

This study addresses the high prevalence of anemia among children under five in Nepal by leveraging the 2022 Nepal Demographic and Health Survey data. To balance model interpretability with the identification of key risk factors, a consensus feature set was constructed through the integration of four feature selection methods: chi-square test, mutual information, point-biserial correlation, and Boruta. The binary classification performance of logistic regression, XGBoost, support vector machines (SVM), deep neural networks (DNN), and TabNet was systematically evaluated on imbalanced data. Results indicate that logistic regression achieved the highest F1-score (0.649) and recall (0.701), SVM attained the best AUC (0.736), and DNN yielded the highest accuracy (0.709), collectively demonstrating the feasibility and potential of machine learning approaches for early screening of childhood anemia.

1 citationsRead paper

CALM: Class-wise Agreement and Label-gated Disagreement Modulation for Decentralized Federated Learning

Sep 05, 2026

Conventional federated learning relies on parameter averaging, which forces clients to be doubly homogeneous: all must run an identical architecture, and accuracy degrades when local data are non-IID. Decentralized federated distillation sidesteps both: each client runs its peers'model snapshots as teachers on its own local data and distills from their soft predictions, with no server, no public data, and no shared architecture. Under severe non-IID skew, however, the trustworthiness of the aggregated teacher target is a matter of degree, yet existing pipelines make hard, all-or-nothing decisions: outlier teachers are discarded by threshold, and whatever target survives is trusted in full. We propose CALM, which replaces every hard decision with a smooth trust gate at three levels: per class, teachers are weighted by agreement with the peer consensus; per sample, distillation is scaled by the teachers'divergence from that target; and a label gate scales it by how strongly the target supports the sample's true label. None of this adds communication or auxiliary data. On CIFAR-10, SVHN, OrganAMNIST, and Google Speech Commands with heterogeneous client architectures under Dirichlet label skew, CALM consistently outperforms uniform and hard-filtered distillation and matches or exceeds competing heterogeneous-FL methods.

0 citationsRead paper
Recent publications

Latest Papers

CALM: Class-wise Agreement and Label-gated Disagreement Modulation for Decentralized Federated Learning

Sep 05, 2026

Conventional federated learning relies on parameter averaging, which forces clients to be doubly homogeneous: all must run an identical architecture, and accuracy degrades when local data are non-IID. Decentralized federated distillation sidesteps both: each client runs its peers'model snapshots as teachers on its own local data and distills from their soft predictions, with no server, no public data, and no shared architecture. Under severe non-IID skew, however, the trustworthiness of the aggregated teacher target is a matter of degree, yet existing pipelines make hard, all-or-nothing decisions: outlier teachers are discarded by threshold, and whatever target survives is trusted in full. We propose CALM, which replaces every hard decision with a smooth trust gate at three levels: per class, teachers are weighted by agreement with the peer consensus; per sample, distillation is scaled by the teachers'divergence from that target; and a label gate scales it by how strongly the target supports the sample's true label. None of this adds communication or auxiliary data. On CIFAR-10, SVHN, OrganAMNIST, and Google Speech Commands with heterogeneous client architectures under Dirichlet label skew, CALM consistently outperforms uniform and hard-filtered distillation and matches or exceeds competing heterogeneous-FL methods.

0 citationsRead paper