Alvaro Coutinho (COPPE/UFRJ/Brazil) and Renan Souza (IBM Research/Brazil) organized a mini-symposium at the Society for Industrial and Applied Mathematics (SIAM) Conference on Computational Science and Engineering. There were eight exciting talks around Scientific Machine Learning (ML) in the Oil and Gas (O&G) Industry. Researchers discussed Physics-informed Neural Network models, provenance of data and ML models and many other relevant topics for this industry.
Souza presented a joint work, in the context of the HPDASC project, between IBM Research, COPPE/UFRJ (Marta Mattoso), and Inria (Patrick Valduriez) on the importance of provenance data management in Scientific Machine Learning [1, 2]. He explained the need for the integration of multiple workflows and the relationship between provenance and domain-specific knowledge to support the training and explainability of O&G ML models. Souza also presented the AI for Seismic Platform, a major project developed by IBM Research Brazil. The official SIAM news website has highlighted this presentation.
[1] R. Souza, L. Azevedo, R. Thiago, E. Soares, M. Nery, M. Netto, E. Brazil, R. Cerqueira, P. Valduriez, and M. Mattoso. Efficient Runtime Capture of Multiworkflow Data Using Provenance. IEEE International Conference on e-Science (eScience), 2019.
[2] R. Souza, L. Azevedo, V. Lourenço, E. Soares, R. Thiago, R. Brandão, D. Civitarese, E. Brazil, M. Moreno, P. Valduriez, M. Mattoso, R. Cerqueira, and M. Netto. Workflow Provenance in the Lifecycle of Scientific Machine Learning. arXiv preprint Databases (cs.DB), 2020.
