About

Bio, education, and research interests of Prof. Adriano Côrtes at UFRJ.

Bio

I am an interdisciplinary researcher with a background spanning Computer Science, Applied Mathematics, and Engineering. I hold a PhD in Computational Mechanics with specialization in High-Performance Computing from the Civil Engineering Program at COPPE/UFRJ, where I conducted research at the Advanced Nucleus for High-Performance Computing (NACAD). I completed a postdoctoral fellowship at the King Abdullah University of Science and Technology (KAUST) in Saudi Arabia.

My academic journey began with a BSc in Computer Science from UFRJ (2004), followed by an MSc in Applied Mathematics from the same institution (2006), building a solid foundation that allows me to navigate across different fields and approach complex problems from multiple perspectives.

Currently, I am Professor Adjunto III at the Federal University of Rio de Janeiro (UFRJ), where I carry out teaching and research activities at the intersection of Computational Science and Engineering, Artificial Intelligence, and High-Performance Computing.

My research focus is on Physics-Informed Machine Learning, where I seek to incorporate physical principles and laws into AI models. I participate in research projects involving diffusion-based generative models and surrogate models using Graph Neural Networks (GNNs). I also have interest in applying AI methods to biological data, including cancer-related projects using single-cell RNA sequencing (scRNA-seq) data.

Education

Year Degree Institution
2013–2015 Postdoctoral Fellowship King Abdullah University of Science and Technology (KAUST)
2009–2013 PhD in Civil Engineering (Computational Mechanics / HPC) COPPE/UFRJ
2008–2009 Lic. in Mathematics UNIGRANRIO
2003–2006 MSc in Applied Mathematics UFRJ
1999–2004 BSc in Computer Science UFRJ

PhD Dissertation

Title: Isogeometric analysis and preconditioning strategies for divergence-conforming spline discretizations for the Stokes problem

Advisor: Prof. Alvaro Luiz Gayoso de Azeredo Coutinho

Keywords: Isogeometric Analysis, Divergence-conforming spline discretization, Block preconditioning strategies

Research Interests

Physics-Informed Machine Learning

Incorporating physical laws — conservation principles, symmetries, governing PDEs — directly into neural network architectures and training procedures. This includes Physics-Informed Neural Networks (PINNs), neural operators (DeepONet, FNO), and hybrid approaches that blend data-driven learning with mechanistic models.

Diffusion-Based Generative Models & Surrogate Modeling

Using probabilistic diffusion models (DDPM) for:

  • Creating digital twins of porous media (2D→3D volumetric translation)
  • Modeling oceanographic systems for deep-water operations
  • Super-resolution techniques to enhance reduced-order model predictions

High-Performance Computing

Scalable parallel algorithms for large-scale scientific simulations. Co-developer of:

  • PetIGA — a high-performance isogeometric analysis framework (140+ citations)
  • PetIGA-MF — multi-field extension for structure-preserving B-spline spaces
  • EdgeCFD — a parallel residual-based variational multiscale code for multiphysics

AI for Biological Data

Applying advanced computational techniques to single-cell RNA sequencing data for cancer research, exploring epigenetic landscape dynamics in glioblastoma multiforme and identifying potential therapeutic targets.

Academic Service

Editorial Boards

  • CNMAC — Proceedings editor, Scientific Computing track (2019–present)
  • Journal of the Brazilian Society of Mechanical Sciences and Engineering (2023–present)

Peer Review

  • Computer Methods in Applied Mechanics and Engineering
  • Journal of Computational Science

Conference Organization

  • Workshop “AI and HPC for Advanced Simulations” — ICCS 2022, 2023, 2024
  • Mini-symposium “Scientific Machine Learning and Uncertainty Quantification” — CILAMCE 2022, 2023, 2024
  • ERAD-RJ 2023 (VIII Escola Regional de Alto Desempenho do Rio de Janeiro)
  • CILAMCE-PANACM 2021 (Local organizing committee)

Contact

  • Email: adriano (arroba) im.ufrj.br (DMA/IM) · adricortes (arroba) cos.ufrj.br (PESC/COPPE)
  • Office: CT — Centro de Tecnologia, Bloco C, Cidade Universitária, Rio de Janeiro, RJ 21941-909
  • Phone: +55 (21) 3938-7396
  • Institutional pages: DMA/IM/UFRJ · PESC/COPPE