Methods for Efficient Coding in Communications and Applications in Machine Learning

• Grant number: 25/22940-2
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: December 01, 2025
• End date: November 30, 2027
• Field of knowledge: Physical Sciences and Mathematics - Mathematics - Applied Mathematics
• Principal Investigator: Sueli Irene Rodrigues Costa
• Grantee: Juliana Gomes Ferreira de Souza
• Host Institution: Instituto de Matemática, Estatística e Computação Científica (IMECC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Company: Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Elétrica e de Computação (FEEC)
• Associated research grant: 20/09838-0 - BI0S - Brazilian Institute of Data Science, AP.PCPE

Abstract

This project aims to analyze and propose efficient coding methods for reliable and secure data transmissions in specific contexts. The focus will be on lattice coding through constructions we have been working on, but other coding methods will also be analysed. Two contexts we are particularly interested, given the perspective of applications in healthcare, are: (i) coding for communication between a central server and user's devices in federated learning and (ii) coding for transmission of sensitive data, where information must be sent securely and with low communication cost. The coding and aggregation algorithms will be analyzed regarding the trade-off between distortion rate and privacy. The results will be validated through simulations in real datasets and comparison with known techniques. The topics of this proposal are in the "Methods" track and with perspective of application in the "Health" track of the BI0S, Brazilian Institute of Data Science Project (FAPESP and MCTIC/CGI - CPA 2020/09838-0) of which the supervisor is one of the main researchers.