EGEdson Satoshi Gomi

Edson Satoshi Gomi

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Edson Satoshi Gomi has a total of 76 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Interpolation of Multivariate Time Series with Nearly-Symmetric Time Translations

    by Bernardo Teixeira, Marcel Barros, Edson Gomi, Eduardo Tannuri, Fábio Levy Siqueira, Anna Helena Reali Costa

  • Pdarts: projected differentiable architecture search for seismic inversion

    by Lucas C. Souza, Carlos G. C. Junior, Ricardo Cerri, Edson S. Gomi, Bruno S. Carmo, Hermes Senger, Murilo Naldi

  • Twinscie-Prov: Gerenciando a Proveniência sobre o Ciclo-de-vida de ML no Sistema Twinscie

    by Júlia Neumann Bastos, Fabio Porto, Fábio Levy Siqueira, Edson Gomi, Ismael Santos, Rodrigo Barreira, Isabela Siqueira, Eduardo Ogasawara

    O aumento da complexidade em aplicações de aprendizado de máquina exige sistemas que garantam rastreabilidade e reprodutibilidade. Este trabalho apresenta a abordagem Twinscie-Prov, uma adequação do padrão W3C PROV para estruturar a proveniência de dados e processos ao longo do ciclo de vida de Machine Learning (ML) no sistema Twinscie. Os dados de proveniência são armazenados no sistema NoSQL Neo4j, permitindo consultas complexas e auditoria. Estudos preliminares mostram que, para consultas envolvendo navegação no grafo de dependências, típicas em dados de proveniência, a implementação no Neo4j é até cinco ordens de grandeza mais rápida que a baseada em logs.

  • Open Access

    Simwave: A finite difference simulator for acoustic waves propagation

    by Hermes Senger, Jaime Freire de Souza, João Baptista Dias Moreira, Keith Jared Roberts, Roussian di Ramos Alves Gaioso, Emílio Carlos Nelli Silva, Edson Satoshi Gomi


  • Open Access

    A Digital Twin System for Oil And Gas Industry: A Use Case on Mooring Lines Integrity Monitoring

    by Vinicius Kreischer de Almeida, Douglas Ericson de Oliveira, Claudio Daniel T. de Barros, Gabriel dos Santos Scatena, Asdrubal N. Queiroz Filho, Fabio Levy Siqueira, Anna Helena Reali Costa, Edson Satoshi Gomi, Leonardo A. F. Mendoza, Evelyn C. S. Batista, et al.

    A Digital Twin is a virtual representation of a real-world object or process, leveraging powerful computational architectures available both on-premises and in the cloud. By harnessing the increased availability of real-time data and advancements in machine learning predictive algorithms, Digital Twins find applications across various domains such as Earth Science, Oil and Gas, and Healthcare. However, realizing their full potential demands addressing the technical complexities of integrating numerous components during development and operational phases of the system. This paper describes an ongoing effort to build a comprehensive platform that supports the entire lifecycle of a Digital Twin, from continuous specialized model training to online prediction and event detection, by capturing and processing live data. This approach enables timely updates to the virtual representations of physical elements within the twin application as they change. We detail each component of the Digital Twin solution and demonstrate its applicability through a real use case implemented in the Oil and Gas industry. Specifically, we focus on monitoring the motion of oil platforms to ensure the integrity of the mooring systems and respond to adverse conditions through an alert system powered by our platform.


  • A Physics-Informed Neural Operator for the Simulation of Surface Waves

    by Marlon Sproesser Mathias, Caio Fabricio Deberaldini Netto, Felipe Marino Moreno, Jefferson Fialho Coelho, Lucas Palmiro de Freitas, Marcel Rodrigues de Barros, Pedro C. de Mello, Marcelo Dottori, Fábio G. Cozman, Anna Helena Reali Costa, et al.

    We develop and implement a neural operator (NOp) to predict the evolution of waves on the surface of water. The NOp uses a graph neural network (GNN) to connect randomly sampled points on the water surface and exchange information between them to make the prediction. Our main contribution is adding physical knowledge to the implementation, which allows the model to be more general and able to be used in domains of different geometries with no retraining. Our implementation also takes advantage of the fact that the governing equations are independent of rotation and translation to make training easier. In this work, the model is trained with data from a single domain with fixed dimensions and evaluated in domains of different dimensions with little impact to performance.


  • Open Access

    Data mapping strategies for multi-GPU implementation of a seismic application

    by Yuri Nicolau Freire, Edson Gomi, Hermes Senger

    Seismic imaging applications are computationally costly, and the industry’s demand is continuously increasing due to the availability of better data, larger data, and the need for better resolution images. It means that the computational capacity needed tends to increase both in terms of FLOPS calculation and memory. Nowadays, many HPC clusters have nodes with multiple GPUs (e.g., 2, 4, and 8). In this paper, we investigate mechanisms and strategies for the data exchange (of the halo zones) of a finite differences grid of a wave simulator implemented in OpenMP. We compare the performance and programming effort of four data mapping mechanisms supported by OpenMP and CUDA. Our best strategy has achieved speedups of 3.87 on four V100 GPUs with NVLink.

  • Embracing Data Irregularities in Multivariate Time Series with Recurrent and Graph Neural Networks

    by Marcel Rodrigues de Barros, Thiago Lizier Rissi, Eduardo Faria Cabrera, Eduardo Aoun Tannuri, Edson Satoshi Gomi, Rodrigo Augusto Barreira, Anna Helena Reali Costa


  • Open Access

    A Physics-Informed Neural Network to Model Port Channels

    by Marlon S. Mathias, Marcel R. de Barros, Jefferson F. Coelho, Lucas P. de Freitas, Felipe M. Moreno, Caio F. D. Netto, Fabio G. Cozman, Anna H. R. Costa, Eduardo A. Tannuri, Edson S. Gomi, et al.

    We describe a Physics-Informed Neural Network (PINN) that simulates the flow induced by the astronomical tide in a synthetic port channel, with dimensions based on the Santos - S\~ao Vicente - Bertioga Estuarine System. PINN models aim to combine the knowledge of physical systems and data-driven machine learning models. This is done by training a neural network to minimize the residuals of the governing equations in sample points. In this work, our flow is governed by the Navier-Stokes equations with some approximations. There are two main novelties in this paper. First, we design our model to assume that the flow is periodic in time, which is not feasible in conventional simulation methods. Second, we evaluate the benefit of resampling the function evaluation points during training, which has a near zero computational cost and has been verified to improve the final model, especially for small batch sizes. Finally, we discuss some limitations of the approximations used in the Navier-Stokes equations regarding the modeling of turbulence and how it interacts with PINNs.

  • Open Access

    Machine learning for noisy multivariate time series classification: a comparison and practical evaluation

    by Aldomar Pietro Santana Silva, Lucas Riera Abbade, Rodrigo da Silva Cunha, Tomaz Maia Suller, Eric Gomes, Edson Satoshi Gomi, Anna Helena Reali Costa

    Multivariate Time Series Classification (MTSC) is a complex problem that has seen great advances in recent years from the application of state-of-the-art machine learning techniques. However, there is still a need for a thorough evaluation of the effect of signal noise in the classification performance of MTSC techniques. To this end, in this paper, we evaluate three current and effective MTSC classifiers – DDTW, ROCKET and InceptionTime – and propose their use in a real-world classification problem: the detection of mooring line failure in offshore platforms. We show that all of them feature state-of-the-art accuracy, with ROCKET presenting very good results, and InceptionTime being marginally more accurate and resilient to noise.