Modern optimization techniques applied to hyperparameter tuning and distance geometry

• Grant number: 24/12967-8
• Support Opportunities: Research Grants - Visiting Researcher Grant - Brazil
• Start date: January 02, 2025
• End date: January 01, 2026
• Field of knowledge: Physical Sciences and Mathematics - Mathematics - Applied Mathematics
• Principal Investigator: Carlile Campos Lavor
• Grantee: Carlile Campos Lavor
• Visiting researcher: Leonardo Delarmelina Secchin
• Visiting researcher institution: Centro Universitário Norte Do Espírito Santo/Ceunes/Ufes, Brazil
• Host Institution: Instituto de Matemática, Estatística e Computação Científica (IMECC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Associated research grant: 23/08706-1 - Numerical optimization, AP.TEM

Abstract

This activity plan focuses on applying continuous optimization techniques to two contexts: hyperparameter tuning in machine learning and distance geometry, specifically in describing the 3D structure of proteins, both of which have significant optimization applications. In addition to theoretical foundations, this project aims to develop and provide computational codes useful to the academic and industrial community. The visit will intensify collaboration between the visiting researcher and the Continuous Optimization Group at IMECC/Unicamp, fostering fruitful research avenues among involved researchers and students. The visiting researcher will also contribute to human resources development by (co-)supervising doctoral theses and master's dissertations in the Applied Mathematics graduate program at IMECC/Unicamp. This research will be conducted within the scope of the thematic project "Métodos Computacionais de Otimização" (FAPESP 2023/08706-1). (AU)