Bridging machine learning and information theory

• Grant number: 18/24912-2
• Support type: Regular Research Grants
• Duration: March 01, 2019 - February 28, 2021
• Field of knowledge: Engineering - Electrical Engineering - Telecommunications
• Cooperation agreement: Imperial College, UK
• Mobility Program: SPRINT - Projetos de pesquisa - Mobilidade
• Principal researcher: Fernando José von Zuben
• Grantee: Fernando José von Zuben
• Principal researcher abroad: Deniz Gündüz
• Institution abroad: Imperial College London, England
• Home Institution: Faculdade de Engenharia Elétrica e de Computação (FEEC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Assoc. researchers: Fernando Rosas ; Henrik Jeldtoft Jensen ; Levy Boccato
• Associated research grant: 13/07559-3 - BRAINN - The Brazilian Institute of Neuroscience and Neurotechnology, AP.CEPID

Abstract

The goal of the proposed project is to enable collaboration between the involved research groups in order to develop i) intelligent communication technologies inspired by modern machine learning (ML) tools, and ii) more efficient and robust learning techniques by exploiting ideas from coding and information theory (IT). The involved researchers have complementary expertise and recent research achievements in the areas of ML and IT, as well as on closely related fields of computational intelligence and communication systems, which will enable immediate cooperation. More specifically, the group aims to investigate practical applications of how ML and IT techniques can be jointly explored for the design of future wireless communication networks that can enable emerging applications of autonomous systems, tactile Internet, and brain-computer interfaces (BCI). In parallel, this project will also explore how ideas from coding and IT can be leveraged to improve the robustness and efficiency of distributed ML algorithms. (AU)