Enhancing Scientific Computation with Machine Learning: Innovations and Breakthroughs in Complex Systems Analysis

• Grant number: 23/15458-4
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: March 01, 2024
• End date: February 28, 2026
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Guido Costa Souza de Araújo
• Grantee: Ronaldo Givisiez Melo Rodrigues
• Host Institution: Instituto de Computação (IC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Associated research grant: 13/08293-7 - CCES - Center for Computational Engineering and Sciences, AP.CEPID

Abstract

This postdoc project is focused on enhancing the capabilities of scientific computation by integrating machine learning (ML) techniques. Our primary objective is to leverage ML algorithms to refine scientific computational models, enabling a better understanding and prediction of complex systems and processes. The proposed work includes designing novel ML-informed algorithms, integrating them into existing scientific computation paradigms, and validating their efficiency and effectiveness in real-world scenarios. By successfully amalgamating machine learning with scientific computation, we anticipate substantial advancements in the precision, reliability, and interpretability of computational models across various scientific disciplines. This includes but is not limited to, molecular dynamics simulations, where the proposed methods could substantially accelerate the prediction of molecular interactions and dynamics; seismic imaging, where ML-informed algorithms can improve the imaging capabilities and interpretation of seismic data; and topology optimization, where the combined approach can lead to more efficient and reliable design solutions. This comprehensive integration of machine learning in scientific computation holds substantial potential for pushing the boundaries of these scientific disciplines.