Zenith  HPDaSc seminar, Monday 7 November 2022, 10h15-11h45
BAT5-01.124, Campus Saint Priest

Dbfication or Dbfiction in Deep Learning Activities
Marta Mattoso
COPPE/UFRJ, Rio de Janeiro, Brazil

Database management techniques have a lot to contribute to generating and selecting a deep learning model. In this talk, we present current initiatives of the database community towards using data management techniques to improve deep learning activities, to discuss pros and cons. Then, we will focus on provenance-based user steering to evaluate different neural network execution configurations. Provenance data adds semantics to the metrics of each configuration, which can help humans in evaluating and reproducing the models proposed by automatic tools.

An introduction to physics-informed neural networks
Alvaro Coutinho
COPPE/UFRJ, Rio de Janeiro, Brazil

In this talk we will give a brief introduction to Physics-Informed Neural Networks (PINN), that are neural networks that encode model equations, like Partial Differential Equations (PDE), as a component of the neural network itself. PINNs are nowadays used to solve PDEs, fractional equations, integral-differential equations, and stochastic problems. PINN fits observed data while reducing a PDE residual. We will discuss the basic math and algorithmic steps and show some current advanced applications.

Journée Zenith, Golfe de Coulondres, 25 May 2022

ML Model Management in Gypscie

Fabio Porto, LNCC, Petropolis, Brazil

To realize the full potential of data science, ML models (or models for short) must be built, combined and ensembled, which can be very complex as there can be many models to select from. Furthermore, they should be shared and reused, in particular, in different execution environments such as HPC or Spark clusters. To address this problem, we propose Gypscie, a new framework that supports the entire ML lifecycle and enables model reuse and import from other frameworks. The approach behind Gypscie is to combine several rich capabilities for model and data management, and model execution, which are typically provided by different tools, in a unique framework. Finally, Gypscie interfaces with multiple execution environments to run ML tasks, e.g., an HPC system such as the Santos Dumont supercomputer at LNCC or a Spark cluster.