8 papers
FedHENet: A Frugal Federated Learning Framework for Heterogeneous Environments
Alejandro Dopico-Castro, Oscar Fontenla-Romero, Bertha Guijarro-Berdiñas +2
Federated Learning (FL) enables collaborative training without centralizing data, essential for privacy compliance in real-world scenarios involving sensitive visual information. M…
Robust gene prioritization for Dietary Restriction via Fast-mRMR Feature Selection techniques
Rubén Fernández-Farelo, Jorge Paz-Ruza, Bertha Guijarro-Berdiñas +2
Gene prioritization (identifying genes potentially associated with a biological process) is increasingly tackled with Artificial Intelligence. However, existing methods struggle wi…
A robust methodology for long-term sustainability evaluation of Machine Learning models
Jorge Paz-Ruza, João Gama, Amparo Alonso-Betanzos +1
Sustainability and efficiency have become essential considerations in the development and deployment of Artificial Intelligence systems, but existing regulatory practices for Green…
CIFNet: An Analytic Neural Learning Framework for Efficient and Calibrated Class-Incremental Learning
Alejandro Dopico-Castro, Oscar Fontenla-Romero, Bertha Guijarro-Berdiñas +1
Class-Incremental Learning (CIL) in deep neural networks is conventionally framed as an iterative gradient-based optimization problem, incurring high computational cost, hyperparam…
Predictively Combatting Toxicity in Health-related Online Discussions through Machine Learning
Jorge Paz-Ruza, Amparo Alonso-Betanzos, Bertha Guijarro-Berdiñas +1
In health-related topics, user toxicity in online discussions frequently becomes a source of social conflict or promotion of dangerous, unscientific behaviour; common approaches fo…
Sustainable techniques to improve Data Quality for training image-based explanatory models for Recommender Systems
Jorge Paz-Ruza, David Esteban-Martínez, Amparo Alonso-Betanzos +1
Visual explanations based on user-uploaded images are an effective and self-contained approach to provide transparency to Recommender Systems (RS), but intrinsic limitations of dat…