Achieving Constraints in Neural Networks: A Stochastic Augmented Lagrangian Approach

no code implementations • 25 Oct 2023 • Diogo Lavado, Cláudia Soares, Alessandra Micheletti

Regularizing Deep Neural Networks (DNNs) is essential for improving generalizability and preventing overfitting.

Low-Resource White-Box Semantic Segmentation of Supporting Towers on 3D Point Clouds via Signature Shape Identification

no code implementations • 13 Jun 2023 • Diogo Lavado, Cláudia Soares, Alessandra Micheletti, Giovanni Bocchi, Alex Coronati, Manuel Silva, Patrizio Frosini

In this paper, we propose SCENE-Net, a low-resource white-box model for 3D point cloud semantic segmentation.

3D Semantic Segmentation Segmentation

Faster than LASER -- Towards Stream Reasoning with Deep Neural Networks

no code implementations • 15 Jun 2021 • João Ferreira, Diogo Lavado, Ricardo Gonçalves, Matthias Knorr, Ludwig Krippahl, João Leite

Whereas reasoning over time-annotated data with background knowledge may be challenging, due to the volume and velocity in which such data is being produced, such complex reasoning is necessary in scenarios where agents need to discover potential problems and this cannot be done with simple stream processing techniques.

Time Series Time Series Forecasting